Relationships Between Fish Size and Otolith Size and Weight in the Bathypelagic Species, Beryx Splendens Lowe, 1834 Collected from the Arabian Sea Coasts of Oman

Quad. Mus. St. Nat. Livorno, 23: 79-84 (2010 ) DOI code: 10.4457/musmed.2010.23.79 79

Relationships between fish size and otolith size and weight in the bathypelagic species, Beryx splendens Lowe, 1834 collected from the Arabian Sea coasts of Oman.

Juma Al-Mamry1, Laith Jawad1,*, Haithem Al-Busaidi1, Saoud Al-Habsi2, Salem Al-Rasbi3

SUMMARY: The regressions between otolith size (length and width), otolith weight and fish length of the benthopelagic species, Beryx splendens living in the Arabian Sea coasts of Oman were provided. No differences between right and left otolith sizes and weight were detected by t-test, so a single linear regression was plotted against standard length (SL) for otolith length (OL), otolith width (OW) and otolith weight (OWE). Data fitted well to the regression model for both OL and OWE to SL (R2 0.6-0.8). These relationships provide a reliable tool in feeding studies and also provide support to palaeontologists in their research on fish fossils. Keywords: benthopelagic fish, otolith, otolith weight, fish-otolith sizes, Beryx, Arabian Sea, Sultanate of Oman.

RIASSUNTO: In questo lavoro viene studiata la regressione tra la forma (lunghezza e larghezza), il peso degli otoliti e la lunghezza della specie bentopelagica Beryx splendens che vive nell’ambiente costiero dell’Oman nel Mare Arabico. Tramite t-test non sono state evidenziate differenze tra la forma e il peso degli otoliti destro e sinistro, è stata registrata una regressione lineare tra la lunghezza standard (SL) e lunghezza (OL), larghezza (OW) e peso (OWE) degli otoliti. I dati coincidono con il modello di regressione sia per OL sia per OWE in rapporto a SL (R2 0.6-0.8). Questo rapporto fornisce un ottimo strumento di supporto per studi sull’alimentazione e per i paleontologi nelle loro ricerche sui pesci fossili. Parole chiave: pesci bentopelagici, otoliti, peso degli otoliti, forma degli otoliti, Beryx, Mare Arabico, Sultanato dell’Oman.

Introduction The high biomass of benthopelagic fish com- munities in all oceans, especially in subtropical Benthopelagic fishes are species usually liv- and tropical seas is an important food resource ing near-bottom environment of the deep-sea in the marine trophic web (Mauchline, Gordon, (Mauchline, Gordon, 1991) and playing an impor- 1991, Bailey et al., 2006). tant ecological role in the transfer of energy from Studies on feeding behaviour have shown the the top surface layer of the sea to the deep en- benthopelagic species to be among the primary vironment. During their diel vertical migration, trophic source for commercially important ben- the benthopelagic fishes feed on micronekton thic fishes (Martin, Christiansen, 1997; Bergstad, and return back to the deeper water layer during Wik, 2003; McIntyre et al., 2006). Moreover, sev- the daytime to avoid predation (Merrett, 1986). eral other predators, such as marine mammals

1. Marine Science Centre, Ministry of Agriculture and Fisheries Wealth, P. O. Box 427, Postal Code 100, Muscat, Sultanate of Oman; 2. Directorate of Fisheries Research, Ministry of Agriculture and Fisheries Wealth, P. O. Box 427, Code 100, Muscat, Sultanate of Oman.; 3. Department of Marine Science and Fisheries, College of Agriculture and Marine Science, Sultan Qaboos University, Muscat, Sultanate of Oman. * Author to whom correspondence should be addressed. Tel. +96899610015; fax: +96824740159; [email protected] 80 Juma Al-Mamry et al. might rely on these food resources (Hassani et Materials and Methods al., 1997; Pauly et al., 1998). The identification and quantification of the Fish specimens were collected during the pe- benthopelagic fish preys are frequently difficult riod 2007-2008 throughout the southern coasts of tasks in the feeding studies (Battaglia et al., 2010). Oman using bottom trawls. The specimens were In general, the prey specimens are already par- identified following Randall (1995). In most cases tially or totally digested and the hard remains the caudal fin was damaged, so standard length in stomachs, intestines, faeces and scats are the (SL; most anterior point to the base of hypural only diagnostic features that can be considered. plate at caudal flexion) in place of total length was Among those hard remains, otoliths are quite considered and measured to the nearest millime- resistant to the digestion and they are considered tre. Sagittae were (total of 132 individuals, i.e., 264 the important tool for prey classification in seve- otoliths) removed through a cut in the cranium to ral dietary studies (Pierce, Boyle, 1991; Pierce et expose them then cleaned and stored dry in glass al., 1991; Granadeiro, Silva, 2000). The examina- vials and the left and right otolith were conside- tion of sagittae from faeces of marine mammals red separately. Specimens with obvious evidence and sea birds usually required to examinme the of calcite crystallization (Strong et al., 1986) or regurgitated digestive pellets in order to identify the preys (Duffy, Laurenso, 1983; Johnstone et al., other aberrant formations were rejected. Each 1990; Pierce, Boyle, 1991). sagittae, systematically placed with the sulcus For the above reasons and for their high acusticus oriented through the observer and its interspecific variability, number of keys and length was determined using hand-held vernier identification guides on fish otolith morphology callipers and defined as the longest dimension has been published (Smale et al., 1995; Campana, between the rostrum and postrostrum axis (no- 2004; Lombarte et al., 2006; Tuset et al., 2008). menclature of Smale et al., 1995) and width as The importance of estimation of the biomass of the dimensions from the dorsal to ventral edge the otolith and their numerical abundance are taken at right angles to the length through the fundamental to understand fish prey’s energy focus of the otolith. Individual otolith weight (in contribution to predator diet. The relationship milligram) was determined using an electronic between fish length and otolith size and weight balance. The relationship between otolith size is usually used in several fish species to build the (length, width) and weight and fish size (SL) were body size and prey biomass. In such a practice, determined using least squares linear regression the otolith measurement has been used (Wyllie for the following parameters: otolith length (OL)- Echeverria, 1987; Gamboa, 1991; Granaderiro, fish length (SL), otolith width (OW)-fish length Silva, 2000; Harvey et al., 2000; Wassle et al., 2003; (SL) and otolith weight (OWE)-fish length (SL). Lychakov et al., 2006; Battaglia et al., 2010). These equations were first calculated for both In paleontological studies, fossil otoliths left and right otoliths and the t-test was used to have proved to be a good taxonomic tool to add check any differences between regressions. The knowledge to the taxonomic status of the ancient regression coefficients were compared and when fauna of the planet when they compared those to significant differences (P<0.05) were not found, the recent reference collections (Nolf 1985, 1995; the H hypothesis (b = b ) was accepted. Girone et al., 2006). Thus, otolith data are widely 0 right left When the equations did not differ statistically, a used in the fields of paleoecology, paleobathy- single linear regression was reported for each pa- metry, paleoclimatology, paleobiogeography and rameter (OL; OW; OWE). The significance of the biostratigraphy (Nolf, 1995). linear regressions was verified using theF -test. The aim of the present study is to provide data about the relationship between otolith size (length and width) and weight and fish length Results of the benthopelagic teleost fish,Beryx splendens Lowe, 1834 collected from the Arabian Sea coasts The range of the standard length of the spe- of Oman. The data offer a helpful tool for feeding cimens used in this study is 130-240 mm with a studies and also give support to palaeontologists mean of 204.1 mm. The fish lengths available for in their investigation on fossils. the species in question were those observed in Relationship between fish size and otolith inBeryx splendens 81 commercial fisheries and research surveys but the al., 2006; Tuset et al., 2008), however, only certain extremes of length ranges were under sampled. geographical areas are covered and the access to Regression of the difference between left and reference material remains requisite (Santos et al., right otolith on fish length indicated slopes not 2001). Thus, an essential objective of researchers significantly different from zero with 0.0048, studying the marine predators’ feeding habits is 0.0129, 0.0805 correlation for otolith length, width to add more information to what is already avai- and weight respectively. Results of regression lable of the fish otolith morphology and on the analysis are given in Table 1. estimation of specific equations, which is useful The range in observed values for otolith leng- to calculate the size and mass of preys. th, width, and weight of the species in question The results of the present paper address to are 9-16, 7-11, and 0.07-0.22 respectively. this need, providing SL-OL, SL-OW and SL-OWE relationships for the benthopelagic fish species, A linear regression model was used to deter- Beryx splendens. Despite the importance role of mine the relationship between the fish length this species in top predators’ diet, its biology and otolith size and weight. Results of regression and ecology had not been well investigated until analyses are given in Table 2. In otolith weight today in Oman at least. and length the regression model appeared to The work at hand presents a preliminary adequately describe the relationship with fish useful tool for better understanding the trophic length with both high correlation (0.7998 and relationships in the Arabian Sea coasts of Oman 0.5901 respectively) and significant estimates for food web. The rebuilding of prey biomass from slopes and intercepts. No observations at lengths otolith size and weight may of benefit to the less than 130 mm were may not be appropriate. benthopelagic fishes of the Arabian Sea Coasts of Oman as to show their role in the bottom of the sea trophic structure, as pointed out by number of feeding studies in several fish groups (Castriotaet al., 2007; Falautano et al., 2007; Consoli et al., 2008; Karakulak et al., 2009). However, up to this day, Tab. 1 - Results of regression analysis of the difference there is a severe lack of data in the Arabian Sea between left and right otolith dimension on fish length. coasts of Oman which hindered an appropriate Tab. 1 - Risultati dell’analisi di regressione della differenza tra le dimensioni degli otoliti destro e sinistro in rapporto alla lun- quantification of their prey biomass and classi- ghezza del pesce. fication to species level in the diet of Arabian Sea coasts of Oman benthic predators. Nothing on record about stomach content of benthic and benthopelagic fishes of the studied area. On contrary of the previous studies on the relationship between fish and sagitta sizes and Tab. 2 - Results of regression analysis of fish length on oto- weight (Wyllie Echeverria, 1987; Gamboa; 1991; lith dimension. SD = standard deviation of the estimated Granadeiro, Silva, 2000; Harvey et al., 2000; Wa- fish length. essle et al., 2003; Battaglia et al., 2010), this paper Tab. 2 - Risultati dell’analisi di regressione della lunghezza del supplies additional information by considering pesce in rapporto alle dimensioni dell’otolite. SD = deviazione standard della lunghezza stimata del pesce. both the otolith length (OL), otolith width (OW) and otolith weight (OWE). It is more suitable Discussion to calculate more than two equations (SL-OL), SL-OW and SL-OWE) since the tip of the otolith Due to their high inter-specific variability rostrum or the dorsal or ventral edges of the in shape, otoliths are considered as a profound otolith may be damaged, making it impossible taxonomic tool in fish species identification (Bat- to measure the OL or OW. Moreover, the coeffi- taglia et al., 2010). Therefore the identification of cient of determination of the SL-OL and SL-OWE fish preys is supported by some reference works linear regression attained a higher value than in (Smale et al., 1995; Campana, 2004; Lombarte et the SL-OW (Table 2). 82 Juma Al-Mamry et al.

In contrast with the findings of Waessle et fish size in some mesopelagic and bathype- al. (2003); Harvey et al. (2000), the otolith of the lagic species from the Mediterranean Sea species in question did not show significant dif- (Strait of Messina), Italy. Scientia Marina, ferences in size between left and right sagittae. 74: 605-612. This finding is in agreement with the results of Bergstad O. A., Wik . D., Hildre Ø., 2003. Preda- Battaglia et al. (2010). tor-Prey relationship and food sources of the In spite of all data fitted well with the linear Sagerrak deep-waterǺ fish assemblage. J. Nor- regressions (SL-OL and SL-OWE), it is advisable thwest Atl. Fisheries Science, 31: 165-180. to use these equations within the fish size ran- Campana S. E., Casselman J. M.,1993. Stock dis- ge limit reported for this species in the results crimination using otolith shape analysis. section. Authors who studied wide range of Canadian Journal of Fisheries and Aquatic fish length and include larvae in their sample, Sciences, 50: 1062-1083. have supplied two different SL-OL regressions, Campana S. E., 2004. Photographic atlas of fish one for the small sized fish and another for the otoliths of the Northwest Atlantic Ocean. Ot- adult specimens (Nishimura, Yamanda, 1988; tawa, Ontario: NRC Research Press pp. 284. Linkowski, 1991). Since the individuals of B. Castriota L., Pipitone C., Campagnuolo S., Ro- splendens collected in the present paper belong to manelli M., Potoschi A., Andaloro F., 2007. the 130-240 mm SL range, the regressions SL-OL Diet of Coryphaena hippurus (Coryphaeni- and SL-OWE calculated here in can be accepted. dae) associated with FADs in the Ionian It is important at this stage to draw the atten- and Southern Tyrrhenian Seas. Cybium, 31: tion to some limitations to the use of biomass 435-441. reconstruction from otolith size and weight. The Consoli P., Romeo T., Battaglia P., Castriota growth of individuals belonging to the same L., Esposito V., Andaloro F., 2008. Feeding species may show som e variations for different habits of the albacore tuna Thunnus alalunga areas and stock (Campana, Casselman, 1993; Rei- (Perciformes, Scombridae) from central chenbacher et al., 2009) or between sexes (Wyllie Mediterranean Sea. Marine Biology, 155: Echeverria, 1987). Furthermore, otolith size might 113-120. become underestimation due to the exposure of Duffy D. C., Laurenson L. J. B., 1983. Pellets of cape cormorants as indicators of diet. Con- the otoliths to chemical and mechanical abrasion dor, 85: 305-307. in the digestive track of predators (Jobling, Brei- Falautano M., Castriota L., Finoia M. G., An- by, 1986; Granadeiro, Silva, 2000). daloro F., 2007. Feeling ecology of little tunny Euthynnus alletteratus in the central Acknowledgements Mediterranean Sea. Journal of the Marine Biological Association of United Kingdom, Our sincere thank goes to the Ministry of 87: 999-1005. Fisheries Wealth, Marine Science and Fisheries Gamboa D. A., 1991. Otolith Size versus Weight Centre and the directorate of Agriculture and and Body-Length Relationships for Eleven Fisheries Developmental Fund for giving us the Fish Species of Baja California, Mexico. Fi- opportunity to work on the fish samples within shery Bulletin U.S., 89: 701-706. the qualitative and quantitative distribution of Girone A., Nolf D., Cappetta H., 2006. Pleistocene marine organisms in Sultanate of Oman and to fish otoliths from the Mediterranean Basin: provide the appropriate financial support. a synthesis. Geobios, 39: 651-671. Granadeiro J. P., Silva. M. A., 2000. The use of References otoliths and vertebrae in the identification and size-estimation of fish in predator-prey Bailey D. M., Ruhl H. A., Smith K. L., 2006. Long- studies. Cybium, 24: 383-393. term change in benthopelagic fish abun- Harvey J. T., Loughlin T. R., Perez M. A., D.S. dance in the abyssal northeast Pacif. Ocean Oxman D. S., 2000. Relationship between Ecology, 87: 549-555. fish size and otolith length for 63 species of Battaglia P., Malara D., Romeo T., Andaloro F., fishes from the eastern North Pacific Ocean. 2010. Relationship between otolith size and NOAA Technical Report NMFS, 150, pp. 35. Relationship between fish size and otolith inBeryx splendens 83

Hassani S., Antoine L., Ridoux V., 1997. Diets of Nishimura A., Yamada J., 1988. Geographical albacore, Thunnus alalunga, and dolphins, differences in early growth of walleye Delphinus delphis and Stenella coerulaeoalba, Pollock Theragra chalcogramma, estimated caught in the northeast Atlantic albacore by back-calculation of otolith daily growth drift-net fishery: A progress report. Journal increments. Marine Biology, 97: 459-465. of the Northwest Atlantic Fisheries Sciences, Nolf D., 1985. Handbook of Paleoichthyology. 22: 119-123. Otolithi piscium. Stuttgart, Springer Verlag. Jobling M., Breiby A., 1986. The use and abuse of 10, pp. 141. fish otoliths in studies of feeding habits of Nolf D., 1995. Studies on fossil otoliths — the marine piscivores. Sarsia, 71: 265-274. state of the art. In: D.H. Secor, J.M. Dean and Johnstone I. G., Harris M. P., Wanless S., Graves S.E. Campana, (eds.). Recent developments J. A., 1990. The usefulness of pellets for as- in fish otolith research, pp. 513-544. Univer- sessing the diet of adult shags Phalacrocorax sity of South Carolina Press. aristotelis. Bird Studies, 37: 5-11. Pauly D., Trites A. W., Capuli E., Christensen V., Karakulak F. S., Salman A., Oray I. K., 2009. Diet 1998. Diet composition and trophic levels of composition of bluefin tuna (Thunnus thyn- marine mammals. ICES J. Marine Science, nus L. 1758) in the Eastern Mediterranean 55: 467-481. Sea, Turkey. Journal of Applied Ichthyology, Pierce G. J., Boyle P. R., 1991. A review of me- 25: 647-651. thods for diet analysis in piscivorous marine Linkowski T. B., 1991. Otolith microstructure and mammals. Oceanogr. Marine Biology Annual growth patterns during the early life histo- Review (London), 29: 409-486. ry of lanternfishes (family Myctophidae). Pierce G. J., Boyle P. R., Diack J. S. W., 1991. Canadian Journal of Zoology, 69: 1777-1792. Identification of fish otoliths and bones in Lombarte A., Chic Ò. Parisi-Baradad V., Olivella faeces and digestive tracks of seals. Journal R., Piera J., García-Ladona E., 2006. A web- of Zoology (London), 224: 320-328. based environment from shape analysis of fish Randall J. E., 1995. Coastal fishes of Oman. otoliths. The AFORO database (http://www.cmi- Bathurst, Australia: Crawford House Publi- ma.csic.es/aforo/). Scientia. Marina, 70: 147-152. shing Pty Ltd., pp. 439. Lychakov D. V., Rebane Y. T., Lombarte A., Fui- Reichenbacher B., Kamrani E., Esmaeili H. R., man L. A., Takabayashi A., 2006. Fish otolith Teimori A., 2009. The endangered cyprin- asymmetry: morphometry and modelling. odont Aphanius ginaonis (Holly, 1929) from Hearing Research, 219: 1-11. southern Iran is a valid species: evidence Martin B., Christiansen B., 1997. Diets and stan- from otolith morphology. Environmental ding stocks of benthopelagic 19 fishes at Biology of Fishes, 86: 504-521. two bathymetrically different midoceanic Santos M. B., Clarke M. R., Pierce G. J., 2001. localities in the northeast 20 Atlantic. Deep- Assessing the importance of cephalopods in Sea Research, 44: 541-558. the diets of marine mammals and other top Mauchline J., Gordon J. D. M., 1991. Oceanic pela- predators: problems and solutions. Fisheries gic prey of benthopelagic fish in the benthic Research, 2: 121-129. boundary layer of a marginal oceanic region. Smale M. J., Watson G., Hecht T., 1995. Otolith Marine Ecology Progress Series, 74: 109-115. atlas of southern African marine fishes. McIntyre J. K., Beauchamp D. A., Mazur M. M., Ichthyological Monograph JLB Smith Insti- Overman N.C., 2006. Ontogenetic trophic tute of Ichthyology, 1: 253. interactions and benthopelagic coupling in Strong M. B., Neilson J., Hunt J., 1986. Aberrant lake Washington: evidence from stable iso- crystallization of Pollock otoliths. Canadian topes and diet analysis. Transaction of the Journal of Fisheries and Aquatic Sciences, American Fisheries Society, 135: 1312-1328. 43: 1457-1463. Merrett N. R., 1986. Biogeography and the ocea- Tuset V. M., Lombarte A., Assis C. A., 2008. Oto- nic rim: a poorly known zone of ichthyofau- lith atlas for the western Mediterranean, north nal interaction. UNESCO Technical Paper of and central eastern Atlantic. Scientia Marina, Marine Science, 49: 201-209. 72S1: 7-198. 84 Juma Al-Mamry et al.

Waessle J. A., Lasta C. A., Bavero M., 2003. Otolith morphology and body size relationships for juvenile Sciaenidae in the Río de la Plata es- tuary (35-36°S). Scientia Marina, 67: 233-240. Wyllie Echeverria T.,1987. Relationship of otolith length to total length in rockfishes from northern and central California. Fishery Bulletin, 85: 383-387.

Riassunto Lungo

Nel tentativo di comprendere meglio le catene trofiche che consentono il sostentamento di pesci commercialmente importanti, l’analisi dei contenuti stomacali risulta particolarmente importante. Attraverso questo tipo di studi, in- fatti, è possibile individuare le specie predate e conoscere le condizioni ecologiche ottimali per il supporto di grandi banchi di pesce pescabile. Sfortunatamente, nella maggior parte dei casi, il contenuto stomacale risulta parzialmente digeri- to e solo poche parti anatomiche possono ancora essere studiate per ottenere le informazioni cer- cate. Tra queste, gli otoliti sono particolarmente importanti perché sono in buona parte resistenti alla digestione e hanno morfologie specie-speci- fiche. Lo studio degli otoliti può dunque fornire informazioni importanti nella ricostruzione delle catene trofiche marine. In questo lavoro viene studiata la regressione tra la forma (lunghezza e larghezza), il peso degli otoliti e la lunghezza della specie bentopelagica Beryx splendens che vive nell’ambiente costiero dell’Oman nel Mare Arabico. Tramite t-test non sono state evidenziate differenze tra la forma e il peso degli otoliti destro e sinistro, è stata registrata una regressione lineare tra la lunghezza standard (SL) e lunghezza (OL), larghezza (OW) e peso (OWE) degli otoliti. I dati coincidono con il modello di regressione sia per OL sia per OWE in rapporto a SL (R2 0.6-0.8). Questo rapporto fornisce un ottimo strumento di supporto per studi sull’alimentazione e per i paleontologi nelle loro ricerche sui pesci fossili.