Relations Between Fecal Pellet Volume and Body Size for Major Zooplankters of the Inland Sea of Japan

Journal of Oceanography Vol. 50, pp. 43 to 49. 1994

Relations between Fecal Pellet Volume and Body Size for Major Zooplankters of the Inland Sea of Japan

SHIN-ICHI UYE and KOHEI KANAME Faculty of Applied Biological Science, Hiroshima University, Higashi-Hiroshima 724, Japan

(Received 5 February 1993; in revised form 24 April 1993; accepted 5 August 1993)

Measurements of fecal pellet volume together with body length/body carbon weight were made for major zooplankters of the Inland Sea of Japan. The pellet volume was highly correlated with animal body size for copepods (10 species combined), a mysid (Neomysis japonica), a larvacean (Oikopleura dioica) and a pelagic shrimp (Acetes japonicus), and a specific equation was given for each group. A single equation could describe the composite relationship between pellet volume (PV, µm3) and body carbon weight (C, µg) for copepods and N. japonica: logPV = 0.85logC + 4.56. Balanid nauplii, O. dioica and a doliolid Dolioletta gegenbauri produced pellets larger, but A. japonicus produced pellets smaller, than those by copepods and N. japonica of equivalent body carbon weight. In general, larger zooplankters produce larger fecal pellets. Hence, the size composition of the zooplankton community is an important parameter for the variation in the vertical flux of material via fecal pellets.

1. Introduction The processes of supply and loss of carbon from the upper mixed layer of the ocean are a central question in Joint Global Ocean Flux Studies (JGOFS). The grazing on phytoplankton and production of fecal pellets are fundamental roles of herbivorous zooplankton in the marine ecosystem. Fecal pellets of zooplankton have been considered the most important vehicle by which the flux out of the surface zone is achieved (Wiebe et al., 1976; Turner and Ferrante, 1979; Bathmann et al., 1987). Both theoretical (Hofmann et al., 1981) and empirical (Sasaki and Nishizawa, 1981; Noji, 1991) studies indicate that large fecal pellets are a prime vehicle for the vertical flux, since larger pellets have higher sinking rate (Smayda, 1969; Fowler and Small, 1972; Small et al., 1979). Since large zooplankters are expected to produce large fecal pellets which sink fast, the zooplankton community consisted of large zooplankters produces high vertical flux of pellets. However, little is known of the relationships between fecal pellet size and zooplankton body size (Shelbourne, 1962; Paffenhöfer and Knowles, 1979). In this study, we investigated the volume of fecal pellets produced by various zooplankters of the Inland Sea of Japan in relation to their body size.

2. Materials and Methods Zooplankters were collected, usually by a NORPAC net (mouth diameter: 0.45 m, length: 2 m, mesh opening: 300 µm, cod-end volume: 1 l), at various parts of the Inland Sea of Japan in 1991. The same type of plankton net with a larger cod-end (volume: ca. 6 l) was used to collect intact specimens of a doliolid Dolioletta gegenbauri and a larvacean Oikopleura dioica occupying a house. A hand net was used for a mysid Neomysis japonica and a pelagic shrimp 44 S. Uye and K. Kaname

Acetes japonicus. Fecal pellets produced from natural diet were obtained by incubating specimens of each species individually for 2Ð24 h in glass vials (volume: 10Ð500 ml, depending on animal size) containing the surface seawater of the sampling sites. In cases where fresh seawater was unavailable, cultured phytoplankton was used as food; Heterosigma akashiwo (Raphidophyceae) was given to copepods of the genus Oithona, and Thalassiosira weissflogii (Bacillariophyceae) was fed to A. japonicus and balanid nauplii. Fecal pellets produced and deposited on the bottom of vials (most larvacean pellets adhered to their house) were collected and their length and width (and depth for D. gegenbauri) were measured under a microscope. For larvaceans which egested few pellets, the pellets remaining in the trunk were measured after being clarified with lactic acid. The volume of pellets was determined assuming that they are cylinder (for copepods, balanid nauplii, mysids, shrimps and larvaceans), spheroid (for copepods, larvaceans and doliolids) or prolate spheroid (for doliolids). Body length (e.g. prosome length for copepods, carapace length for shrimps, trunk length for larvaceans and standard body length for the others, see Uye, 1982) were measured and converted to body carbon weight using regression equations (Paffenhöfer, 1976, for O. dioica; Hirota, 1981, for D. gegenbauri; Uye, 1982, 1988, 1991a, unpublished, for copepods, balanid nauplii, N. japonica and A. japonicus) and conversion factors from dry weight to carbon weight (Uye, 1982, for O. dioica; Madin et al., 1981, for D. gegenbauri).

3. Results

3.1 Copepods The ratio of length/width of fecal pellets varied from 2.2 to 13.0 for copepods, which included 10 species (Acartia omorii, Calanus sinicus, Centropages abdominalis, Microsetella norvegica, Oithona davisae, O. similis, Oncaea media, Paracalanus sp., Pseudodiaptomus marinus, Sinocalanus tenellus). Mean pellet volume is plotted against prosome length (Fig. 1) and against body carbon weight (Fig. 2). C. sinicus, the largest copepod species in this study,

Fig. 1. Relationship between mean fecal pellet volume and prosome length for copepods. Zooplankton’s Fecal Pellet Volume-Body Size Relationships 45

Fig. 2. Relationship between mean fecal pellet volume and body carbon weight for copepods. Pellet sinking rate is also shown from the relationship between sinking rate and pellet volume (Small et al., 1979). produced the largest pellets (3.26 × 106 µm3), while O. davisae, the smallest species, laid the smallest ones (3.72 × 103 µm3). The pellet volume was highly correlated with prosome length:

logPV = 2.58logPL Ð 2.38 (N = 99, r = 0.97) as well as with body carbon weight:

logPV = 0.90logC + 4.55 (r = 0.95) where PV is mean volume of a fecal pellet (µm3), PL is prosome length (µm) and C is body carbon weight (µg). However, there was a considerable variation in pellet volume within a species with equal body size. For example, C. sinicus adult females with prosome length of ca. 2300 µm produced pellets with volume ranging from 4.02 × 105 to 3.26 × 106 µm3 (8-fold of difference). This individual variation might not be caused solely by the difference of food condition, since the variation was observed for animals which had been collected at the same location and incubated under the same condition. An equation by Small et al. (1979), which describes the relationship between sinking rate (S, m dayÐ1) and volume of pellets (PV, µm3) for copepods and euphausiids:

logS = 0.513logPV Ð 1.214 is used to estimate the sinking rate of copepod pellets obtained in the present study (Fig. 2). The sinking rate is as high as ca. 100 m dÐ1 for pellets of C. sinicus, but is only ca. 4 m dÐ1 for those of O. davisae.

3.2 Neomysis japonica The width of fecal pellets produced by a given individual was constant, but the length was 46 S. Uye and K. Kaname variable. The length/width ratios varied from 11.6 to 34.3. The mean pellet volume (PV, µm3) of N. japonica was significantly related to body length (BL, mm):

logPV = 1.98logBL + 4.88 (N = 20, r = 0.79) and to body carbon weight (C, µg) (Fig. 3):

logPV = 0.64logC + 4.98 (r = 0.79).

3.3 Acetes japonicus The pellets of A. japonicus were string-like. Since some of them were apparently broken apart, the length/width ratios varied as wide as from 1.6 to 69.6. The relation of carapace length (CL, mm) to mean pellet volume (PV, µm3) was:

logPV = 1.67logCL + 5.79 (N = 19, r = 0.70) and the relation of body carbon weight (C, µg) to mean pellet volume was (Fig. 3):

logPV = 0.79logC + 4.05 (r = 0.72).

3.4 Oikopleura dioica The length/width ratios of the larvacean pellets varied from 1.7 to 5.3. The relationship between mean pellet volume (PV, µm3) and trunk length (TL, µm) was:

logPV = 2.83logTL Ð 2.54 (N = 58, r = 0.79)

Fig. 3. Relationships between mean fecal pellet volume and body carbon weight for various taxonomic groups of zooplankton. Regression lines for Neomysis japonica (2), Acetes japonicus (3) and a larvacean Oikopleura dioica (4) are shown in addition to copepods (1). Regression equations are shown in text. Zooplankton’s Fecal Pellet Volume-Body Size Relationships 47 and the relationship between mean pellet volume and body carbon weight (C, µg) was (Fig. 3):

logPV = 1.14logC + 4.91 (r = 0.79).

3.5 Other zooplankton There were insufficient data for balanid nauplii (N = 4) and D. gegenbauri (N = 5) to de- termine the specific relationship between pellet volume and body size. Mean pellet volume of balanid nauplii ranged from 2.3 × 104 to 5.5 × 104 µm3. Mean pellet volume of D. gegenbauri ranged from 1.5 × 106 to 5.1 × 106 µm3.

3.6 Overall relations Mean pellet volume is plotted against individual body carbon weight for each taxonomic group (Fig. 3). In combination of the data for copepods and N. japonica, a highly significant regression equation can be established between pellet volume (µm3) and body carbon weight (C, µg):

logPV = 0.85logC + 4.56 (N = 119, r = 0.93).

Fecal pellets produced by balanid nauplii, O. dioica and D. gegenbauri are larger than those produced by copepods and N. japonica of equivalent body carbon weight. However, A. japonicus produce small fecal pellets relative to those of copepods and N. japonica.

4. Discussion Herbivorous metazoan zooplankters graze phytoplankton cells, smash them into small pieces with teeth and pelletize them in a digestive tract, and the volume of their fecal pellets is generally larger than that of food particles. Protozoan zooplankters, however, lack a digestive tract, and their fecal particles are the same or smaller than food particles (Stoecker, 1984). Hence, metazoan zooplankters contribute to an acceleration, while protozoan zooplankters contribute to a reduction, of the loss of materials from the surface ocean. The contribution of fecal pellets to the benthos was investigated by Hofmann et al. (1981) through an analytic model which considers pellet production by different size (stage) groups of a copepod, e.g. Paracalanus. According to the model, the concentration and vertical flux of pellets is a function of producer size, consumer size and abundance. Large zooplankters produce large pellets which sink fast, decompose slowly and are eaten by few feeders because of size. On the other hand, small zooplankters produce small pellets which sink slowly and are subjected to higher decomposition and feeding rates, resulting in low vertical flux of pellets. Hence, the size composition of the zooplankton community is a prime factor in the variation of the vertical flux of pellets. The size composition of the zooplankton community varies geographically as well as seasonally. It is a general concept that small zooplankters are relatively important in tropical and subtropical oceans, while large zooplankters are dominant in temperate and boreal waters (Beers and Stewart, 1971; Endo et al., 1983). There are also geographical changes in size composition of the zooplankton community along an inshore-offshore transect on the continental shelf (Uye et al., 1990). Uye et al. (1990) investigated the size composition of the copepod community along a transect from the innermost part of Osaka Bay to Kii Channel, the Inland Sea of Japan. The copepod community consisted mainly of the small species, Paracalanus sp., Corycaeus spp. and 48 S. Uye and K. Kaname

Microsetella norvegica in the inner Osaka Bay, and was dominated by the large species Calanus sinicus in Kii Channel. In June, the regional difference in the median weight increased more than 80-fold, which corresponds to more than 52-fold of difference in pellet volume and to more than 8-fold of difference in sinking rate. From these facts, it can be surmised that the vertical flux of fecal pellets is much higher in Kii Channel than in inner Osaka Bay, if the copepod community egestion rate is equal. In Tokyo Bay, the copepod community consisted monospecifically of the small species O. davisae in summer (Uye, 1991b). In this bay, the vertical flux of pellets can be much reduced in comparison to the flux in Osaka Bay. Although the pellets produced by D. gegenbauri are larger than those by copepods of equivalent carbon weight, the former is apparently less compacted than the latter. Hence, the sinking rate of the pellets of D. gegenbauri is lower than is expected based on sinking rate of copepod pellets (Deibel, 1990). Deibel (1990) also found that most of the pellets produced by specimens fed only naturally occurring particles do not sink, while the pellets produced from amended food of cultured phytoplankton do sink. This means that the variability in the relationship between sinking rate and volume of fecal pellets is resulted from the different pellet composition and compaction. Most of pellets produced by O. dioica adhere to the inside wall of the gelatinous house. As many as 38 pellets were observed in a house. Accumulation of pellets within discarded houses may enhance the vertical flux of larvacean pellets comparing to sinking of individual pellets. High sedimentation rate of fecal pellets of Oikopleura longicauda (annual av.: 66% of total zooplankton pellets in terms of volume) was reported by Taguchi (1982) in Kaneohe Bay, Hawaii. Many processes may work on fecal pellets during sedimentation in the water column (e.g. mechanical destruction by hydrodynamics, aggregation, bacterial decomposition, destruction and ingestion by zooplankton). Although the actual sedimentation rate is much reduced than the egestion rate by the zooplankton community in the surface zone (Lampitt et al., 1990; Noji, 1991), the size of pellets at a time of freshly released is primary important to estimate the vertical flux of pellets. Although there are considerable intraspecific variations due to difference of food condition (Dagg and Walser, 1986), the present study demonstrates that the volume of pellets can be estimated from individual body carbon weight for each taxonomic group. The regressions obtained for zooplankters of the Inland Sea of Japan may be helpful to predict the relationship between pellet volume and zooplankton body size in other waters. If the egestion rate of the zooplankton is known, the size composition of fecal pellets at a time of freshly produced can be estimated by only knowing the size composition of the major taxonomic groups of the zooplankton community.

Acknowledgements We would like to thank Dr. T. Onbé for the helpful comments. Thanks are also due to the captain and crew of T&R.V. Toyoshio-Maru for their aid in sampling at sea.

References Bathmann, U. V., T. T. Noji, M. Voss and R. Peinert (1987): Copepod fecal pellets: abundance, sedimentation and content at a permanent station in the Norwegian Sea in May/June 1986. Mar. Ecol. Prog. Ser., 38, 45Ð51. Beers, J. R. and G. L. Stewart (1971): Micro-zooplankters in the plankton communities of the upper waters of the eastern tropical Pacific. Deep-Sea Res., 18, 861Ð883. Dagg, M. J. and W. E. Walser, Jr. (1986): The effect of food concentration on fecal pellet size in marine copepods. Limnol. Oceanogr., 31, 1066Ð1071. Zooplankton’s Fecal Pellet Volume-Body Size Relationships 49

Deibel, D. (1990): Still-water sinking velocity of fecal material from the pelagic tunicate Dolioletta gegenbauri. Mar. Ecol. Prog. Ser., 62, 55Ð60. Endo, Y., H. Hasumoto and A. Taniguchi (1983): Microzooplankton standing crop in the western subtropical Pacific off the Bonin Islands in winter, 1980. J. Oceanogr. Soc. Japan, 39, 185Ð191. Fowler, S. W. and L. F. Small (1972): Sinking rates of euphausiid fecal pellets. Limnol. Oceanogr., 17, 293Ð296. Hirota, R. (1981): Methods for netzooplankton investigation. p. 177Ð191. In Manual for the Investigations of the Coastal Marine Environment, ed. by the Oceanogr. Soc. Japan, Koseisha Koseikaku, Tokyo (in Japanese). Hofmann, E. E., J. M. Klinck and G.-A. Paffenhöfer (1981): Concentrations and vertical fluxes of zooplankton fecal pellets on a continental shelf. Mar. Biol., 61, 327Ð335. Lampitt, R. S., T. Noji and B. von Bodungen (1990): What happens to zooplankton faecal pellets? Implications for material flux. Mar. Biol., 104, 15Ð23. Madin, L. P., C. M. Cetta and V. L. McAlister (1981): Elemental and biochemical composition of salps (Tunicata: Thaliacea). Mar. Biol., 63, 217Ð226. Noji, T. T. (1991): The influence of macrozooplankton on vertical particulate flux. Sarsia, 76, 1Ð9. Paffenhöfer, G.-A. (1976): On the biology of Appendicularia of the southeastern North Sea. p. 437–455. In Proc. 10th Europ. Symp. Mar. Biol.. Belgium, Vol. 12, ed. by G. Persoone and E. Jaspers, Universa Press, Wetterren. Paffenhöfer, G.-A. and S. C. Knowles (1979): Ecological implications of fecal pellet size, production and con- sumption by copepods. J. Mar. Res., 37, 35Ð49. Sasaki, H. and S. Nishizawa (1981): Vertical flux profiles of particulate material in the sea off Sanriku. Mar. Ecol. Prog. Ser., 6, 191Ð201. Shelbourne, J. E. (1962): A predator-prey relationship for plaice larvae feeding on Oikopleura. J. Mar. Biol. Ass. U.K., 42, 243Ð252. Small, L. F., S. W. Fowler and M. Y. Unlu (1979): Sinking rates of natural copepod fecal pellets. Mar. Biol., 51, 233Ð 241. Smayda, T. J. (1969): Some measurements of the sinking rate of fecal pellets. Limnol. Oceanogr., 14, 621Ð625. Stoecker, D. K. (1984): Particle production by planktonic ciliates. Limnol. Oceanogr., 29, 930Ð940. Taguchi, S. (1982): Seasonal study of fecal pellets and discarded houses of Appendicularia in a subtropical inlet, Kaneohe Bay, Hawaii. Estuar. Coast. Shelf Sci., 14, 545Ð555. Turner, J. T. and J. G. Ferrante (1979): Zooplankton fecal pellets in aquatic ecosystems. Bioscience, 29, 670Ð677. Uye, S. (1982): Length-weight relationships of important zooplankton from the Inland Sea of Japan. J. Oceanogr. Soc. Japan, 38, 149Ð158. Uye, S. (1988): Temperature-dependent development and growth of Calanus sinicus (Copepoda: Calanoida) in the laboratory. Hydrobiologia, 167/168, 285Ð293. Uye, S. (1991a): Temperature-dependent development and growth of the planktonic copepod Paracalanus sp. in the laboratory. Bull. Plankton Soc. Japan, Spec. Vol., 627Ð636. Uye, S. (1991b): Material transport in Tokyo Bay—The roles of small zooplankton. Kaiyo Monthly, 23, 194Ð198 (in Japanese). Uye, S., K. Imaizumi and O. Matsuda (1990): Size composition and estimated phosphorus regeneration rates of the copepod community along an estuarine-offshore transect in the Inland Sea of Japan. Estuar. Coast. Shelf Sci., 31, 851Ð863. Wiebe, P. H., S. H. Boyd and C. Winget (1976): Particulate matter sinking to the deep-sea floor at 2000 m in the Tongue of the Ocean, Bahamas, with a description of a new sedimentation trap. J. Mar. Res., 34, 341Ð354.