Feeding ecology of the sand shrimp Crangon hakodatei Rathbun, 1902 (Decapoda: Crangonidae) in the East Sea of Korea

Abstract

The feeding habits of the sand shrimp Crangon hakodatei in the East Sea (Korea) were investigated through analysis and comparison of the stomach contents of 602 individuals according to season, shrimp size class, and prey diversity. The diet of C. hakodatei consisted of 17 prey categories mainly comprising crustaceans, molluscs, polychaetes, nematodes, algae, and fishes, with crustaceans dominating the diet. Molluscs, nematodes, and fishes were also important prey, whilst other categories including polychaetes and algae comprised small percentages of the diet. For small C. hakodatei individuals (<10 mm carapace length [CL]), amphipods and mysids comprised more than 67% of the prey in both relative abundance and frequency of occurrence. Large individuals (>10 mm CL) tended to be more dependent on amphipods than mysids. Amphipods and mysids together constituted the dominant prey, accounting for more than 50% of the diet in terms of both percent occurrence and relative abundance. The abundance and occurrence of prey showed a seasonal variation, with amphipods and mysids being the predominant prey in autumn (45%), winter (30%), and spring (40%). Amphipods were the dominant prey with regard to season, size class, sex, and area.

Keywords:

• Crangon hakodatei
• prey items
• feeding habits
• amphipods
• mysids

Introduction

The sand shrimp Crangon hakodatei Rathbun, 1902 is an abundant species in the East and Yellow seas (Hayashi & Kim 1999; Cha et al. 2001), occurring on sandy, muddy, or mixed substrata in sublittoral areas from 10 to 250 m depth. C. hakodatei is differentiated from other related species by the high blunt median dorsal carina on the third to fifth abdominal somites. Crangon nigricauda, reported from Korea by Kim and Park (1972), was later reidentified as C. hakodatei by Kim (1976). Crangon affinis, which was reported from northern China by Liu (1955), is referred to as C. hakodatei by Hayashi and Kim (1999) and Chen et al. (2006).

As with other cardiean shrimps, Crangon spp. play an important role in the trophic processing of nekton (Chuhukalo & Shebanova 2008). They are commonly distributed on sandy, muddy, or mixed benthic substrata of littoral and sublittoral areas from cold to temperate regions of the northern hemisphere. The sand shrimp is one of the most abundant and important components of estuarine and coastal soft bottom habitats (where it is both prey and predator in the associated communities) and also has high commercial value as a protein source for human consumption (Price 1962; Tiews 1970; Boddeke 1989; Hayashi & Kim 1999; Oh et al. 2001; Hanamura & Matsuoka 2003).

A number of studies have investigated the feeding behavior of crangonid shrimp including C. affinis by Kosaka (1970) and Hong and Oh (1989); C. allmani by Allen (1960); C. crangon by Lloyd and Yonge (1947); Tiews (1970), and Oh et al. (2001); C. franciscorum and C. nigricauda by Wahle (1985); C. septemspinosa by Price (1962) and Wilcox and Jeffries (1974); and C. uritai by Nakaya et al. (2004). However, despite its ecological and economic importance, there is no information on the diet of C. hakodatei. The present study was therefore undertaken to describe the diet composition of this species. The results increase our understanding of the feeding strategies and patterns of C. hakodatei and also provide information on the relationships between this and other species and their effect in marine ecosystems. Such information will facilitate estimates of the energy flow from primary producers to higher trophic levels. Trophic interactions are central to establishing the ecological role of species in ecosystem communities, and stomach content analyses can enable descriptions of the composition of the diet of species and assignment of their trophic level (Lorman & Magnuson 1978; Momot et al. 1978). Estimation of prey abundance and frequency of occurrence, and the relative importance of food sources, can provide critical ecological information (Espinoza & Wehrtmann 2008) and information on the composition of various food types (Crisp 1963; Fagade & Olaniyan 1972).

The objective of this study was to investigate the diet and feeding behavior of C. hakodatei. The specific objectives were to (1) obtain data on the feeding ecology of C. hakodatei; (2) investigate seasonal variations in diet and size differentiation in its feeding habits; and (3) derive information on the relationship of this species to its prey and its prey preferences. All animals are selective with respect to their prey (Krebs 1989). These objectives were addressed using stomach content analysis and laboratory and field observations.

Materials and methods

Sampling

The C. hakodatei samples were obtained from trawls conducted in three areas (areas 1–3) from the Gijang coast, South Korea (Figure 1). All individuals were collected randomly from depths of 30 to 60 m in each area using a small bottom trawl. The specimens of C. hakodatei were fixed for 24 hours in 4% neutralized formalin and stored in 70% alcohol. Samples were collected from area 3 during the period September 2010 to June 2011 and from areas 1 and 2 during the period from August 2010 to October 2010. The specimens from area 3 were grouped according to season: autumn (September–November), winter (December–February), and spring (March–May).

Figure 1.  Sampling areas in this study, areas 1, 2, and 3 are those of rectangle, circle, and triangle, respectively.

Laboratory analysis

The carapace length (CL; the shortest distance between the posterior margin of the orbit and the mid dorsal posterior edge of the carapace) of each C. hakodatei specimen was measured using Vernier calipers (accuracy 0.01 mm). The specimens were grouped into two size classes: small (<10 mm CL) and large (>10 mm CL). The sex of specimens was determined by binocular microscope examination of the endopod of the second pleopods and by the presence of the male appendage.

The relative degree of stomach fullness was assessed visually, and each foregut was assigned to one of five categories derived from the points method described by Wear and Haddon (1987). The fullness categories and points were as follows: (1) full: 95%–100% full, 100 points; (2) semi full: <95% to >65% full, 75 points; (3) half full: <65% to >35% full, 50 points; (4) semi empty: <35% to >5% full, 25 points; and (5) empty: 5% or less full, 2.5 points. Prey items present in the stomach contents were identified to the lowest taxonomic level possible. Specimens with more than 2.5 points were excluded from the analyses, as were sand and mud. Prey was determined as being either present or absent and as a proportion of the number of points assigned for stomach fullness. Diet was examined for three months in areas 1 and 2 and for three seasons in area 3 (autumn, winter, and spring).

Each stomach was removed and opened on a microscope slide. The contents were first examined at low magnification using a dissecting microscope and then examined again at high magnification using a light microscope. Apparently, empty stomachs were also examined. Prey items from the stomach contents were identified using the following criteria: (1) Polychaetes: brownish or yellowish stout, tapering iridescent chaetae, stout dark brown jaws; (2) Fishes: whole bones or pieces of rib, vertebrae, muscle tissue, scales, or eggs; (3) Crustaceans: fragments of chitinous shell, parts of legs or claws, or parts of the telson; (4) Molluscs (gastropods/bivalves): parts of shell, part of the helical coil; and (5) Algae (diatoms): circular blue–green/gray items or items having radial iridescence. Sand and mud particles were found among food items in the stomach contents, as has been observed in studies of other marine animals. In menhaden deposit feeders, the diet consists of a combination of sediment grains (sand, mud), adsorbed material, and detritus comprising plant and algal material (Deegan et al. 1990). Land-derived organics have also been found (Conkright & Sackett 1986).

Data and statistical analysis

Many indices have been used to describe the importance of various preys in the diet of fish (Hynes 1950; Hyslop 1980). In the present study, percent frequency of occurrence (F) and relative abundance (A) of each prey type were estimated using the following formulae:

where n _i is the number of shrimp with prey i in the stomach contents, N is the total number of shrimp having stomach contents, S _i is the number of prey i, and S _t is the total number of prey items.

Trophic diversity (H′) in the diet (with respect to season and size class) was calculated according to the following equation (Shannon–Wiener index; Cody & Diamond 1975):

where S is the total number of species and p _i is the frequency of the ith species.

Diet equality was calculated for the various size classes and seasons, using Pielou's (1975) evenness index according to the following equation:

where H′ is the number derived from the Shannon diversity index, and H′ _max is the maximum value of H′.

The data obtained were analyzed using the Kruskal–Wallis test to compare differences in stomach fullness among sizes and season, according to the equation of Conover (1971) and Daniel (1974):

This test indicates whether there is a difference, but provides no information on the cause of any difference.

To test whether the observed frequencies among feeding parameters differed from expected values, we used the χ² (chi-squared) test. Statistical analyses were undertaken using SPSS Statistics version 19 and SAS version 9.1.3.

Results

Size composition

During the study, 602 individuals of C. hakodatei were collected and analyzed. Amongst these, the 304 individuals from areas 1 and 2 were used to assess the effect of area on feeding habits, and the 298 individuals from area 3 were used to assess the effect of seasonal variation on feeding habitats. Of all the stomachs that were examined from area 1 and 2, about 12% of stomachs from males and about 9% of those from females were found to be empty, while the proportion of empty stomach for males and females in area 3 comprised 10% and 11%, respectively. The length of specimens in the study ranged from 6.24 to 15.8 mm CL. No significant difference in the size distribution among areas was found (Kolmogorov–Smirnov 2-sample test, p=0.993), and there were no significant differences in size distribution among seasons (Kolmogorov–Smirnov 2-sample test; autumn and winter, p=0.808; autumn and spring, p=0.641; winter and spring, p=0.993).

Stomach fullness

The relative degree of stomach fullness was assessed visually and each foregut was assigned to one of five categories (100, 75, 50, 25, and 2.5points). Of the stomachs examined among seasons (area 3), approximately 88% received scores greater than 2.5 points on the stomach fullness scale. There was no significant difference with respect to the stomach fullness proportion and season between males and females or for each size class (Kruskal–Wallis test, p>0.05) (Figure 2).

Figure 2.  Proportion of stomach fullness scale of Crangon hakodatei by size for (A) autumn, (B) winter, and (C) spring.

Diet composition

Mysids and amphipods were the most important food items with respect to both area (areas 1 and 2) and season (area 3), based on the percent frequency of occurrence and relative abundance (Table 1 and Figure 3). The major components of the stomach contents of C. hakodatei were polychaetes, crustaceans (mysids, amphipods, isopods, decapods, and copepods), mollusc shells (primarily gastropods and bivalves), fish vertebrae, nematodes, and algae. These prey categories comprised 89.1% of the relative abundance and 92.79% of total occurrence for males and 85.9% of the relative abundance and 91.32% of the total occurrence in females. Mysids and amphipods were the dominate food items overall, both being present in each of the foreguts examined.

Figure 3.  Relative importance of major stomach contents of Crangon hakodatei in overall diets for all combined samples irrespective of areas (Alg, algae; Amp, amphipods; Cum, cumaceans; Dec, decapods; Iso, isopods; Mol, molluscs; Mys, mysids; Pis, pisces; Pol, polychaets; Otc, other crustaceans; Oth, other species; Nem, nematodes).

Table 1. Diet composition of Crangon hakodatei in the three seasonal groups and two size groups (%F, frequency of occurrence; %N, percentage abundance).

Area	Autumn				Winter				Spring
Size class	Small		Large		Small		Large		Small		Large
Number examined	47		52		42		42		40		41
Prey item	%F	%N	%F	%N	%F	%N	%F	%N	%F	%N	%F	%N
POLYCHAETA	2.13	1.00	9.62	5.38	9.52	6.25	2.38	1.23	22.50	12.00	7.32	1.70
CRUSTACEA
Mysidacea	23.40	18.00	21.15	17.20	26.19	16.30	26.19	18.50	22.50	20.00	12.20	13.00
Amphipoda	38.30	22.00	32.69	21.50	38.10	27.50	33.33	23.50	42.50	24.00	31.71	16.00
Isopoda	6.38	5.00	9.62	5.38	7.14	3.75	2.38	1.23	0.00	0.00	7.32	1.00
Cumacea	8.51	5.00	1.92	1.08	7.14	5.00	4.76	4.94	7.50	4.00	19.51	3.20
Decapoda	2.13	1.00	5.77	3.23	7.14	3.75	9.52	7.41	15.00	8.00	4.88	4.00
Copepoda	4.26	3.00	9.62	7.53	2.38	1.25	9.52	6.17	2.50	1.80	2.44	5.00
Unidentified pieces	14.89	8.00	3.85	3.23	9.52	6.25	7.14	3.70	7.50	4.00	4.88	2.10
Other crustacean	4.26	3.00	9.62	6.45	7.14	3.75	7.14	3.70	7.50	4.00	12.20	3.00
MOLLUSCA
Bivalvia	2.13	1.00	5.77	4.30	7.14	3.75	4.76	2.47	7.50	3.20	7.32	6.00
Gastropoda	6.38	4.00	1.92	1.08	4.76	2.50	7.14	4.94	7.50	4.00	7.32	6.00
FISH
Pisces	6.38	4.00	1.92	1.08	7.14	3.75	4.76	2.47	0.00	0.00	0.00	0.00
Backbone	4.26	3.00	1.92	1.08	0.00	0.00	7.14	4.94	7.50	4.00	4.88	4.60
Eggs	6.38	2.00	5.77	3.23	4.76	2.50	0.00	0.00	7.50	4.00	2.44	0.40
ALGAE	6.38	4.00	3.85	2.15	0.00	0.00	0.00	0.00	2.50	2.80	0.00	0.00
OTHER	19.15	3.00	19.23	14.00	14.29	7.50	19.05	12.30	15.00	8.00	7.32	5.20
NEMATODA	0.00	13.00	3.85	2.15	11.90	6.25	4.76	2.47	0.00	0.00	7.32	0.00

Diet differences by size class

The diet composition for shrimp in the two size classes (small and large) in areas 1 and 2 is shown in Figure 4. For those in the small size class, amphipods were clearly dominant in each area in terms of both relative abundance and frequency of occurrence; they comprised more than 39% and 30% of the prey items in areas 1 and 2, respectively. For shrimp in the large size class, mysids and amphipods were dominant prey in both areas. In both relative abundance and frequency of occurrence, mysids and amphipods comprised more than 65% of prey items in area 1 and more than 70% in area 2. No significant differences were found in the proportion of prey consumed by the two size classes in each area (χ² test, df=13, p>0.05).

Figure 4.  Relative importance of major stomach content items for areas 1 and 2 samples pooled by size classes: small (<10 mm CL) and large size (>10 mm CL) (Alg, algae; Amp, amphipods; Cum, cumaceans; Dec, decapods; Iso, isopods; Mol, molluscs; Mys, mysids; Pis, pisces; Pol, polychaets; Otc, other crustaceans; Oth, other species; Nem, nematodes).

Seasonal differences in diet

The relative abundance and frequency of occurrence of food items in stomach contents showed seasonal fluctuations (Figure 5). Mysids and amphipods were the predominant prey in all three seasons, and based on relative abundance and frequency of occurrence, in combination accounted for 40% of the diet in autumn, 30% in winter, and 41% in spring. Mysids and amphipods were also the predominant prey items for each size class in all seasons. Each gender showed a mixed feeding strategy, with varying degrees of specialization and generalization on different prey types, but in terms of relative abundance and frequency of occurrence, the diets of males and females were similar, including among seasons: mysids and amphipods dominated the diet in autumn (45% males, 40% females), winter (41% males, 32% females), and spring (43% males, 39% females).

Figure 5.  Relative importance of major stomach contents for Crangon hakodatei samples pooled by seasons (A) autumn, (B) winter and (C) spring (Alg, algae; Amp, amphipods; Cum, cumaceans; Dec, decapods; Iso, isopods; Mol, molluscs; Mys, mysids; Pis, pisces; Pol, polychaets; Otc, other crustaceans; Oth, other species; Nem, nematodes).

Trophic diversity and equality

Seasonally trophic diversity was generally low (Figure 6A). For both size classes, diversity was highest in autumn. No appreciable differences were evident between size classes in any season. Trends were similar for diet evenness (Figure 6B). The higher index values in autumn indicated that the prey items consumed were more evenly distributed, as demonstrated by the relative importance of the seasonal diet composition. In other seasons, the shrimps were more dependent on a small number of prey items.

Figure 6.  Trophic diversity (A) and equality (B) of prey items found in small and large size classes of Crangon hakodatei at each season.

Discussion

The most important prey items in the C. hakodatei diet were crustaceans, with other food classes not playing an important role. The dominance of crustaceans in the diet of several species of Crangon has been reported, including C. crangon (Pihl & Rosenberg 1984; Oh et al. 2001), C. franciscorum (Sitts & Knight 1979), and C. uritai (Hanamura & Matsuoka 2003).The diet of C. hakodatei consists of benthic organisms in three main categories: (1) organisms that, as a result of vertical migration, occur close to the benthos during part of the day (e.g. mysids, shrimps, fishes); (2) organisms that occur on or just beneath the surface of the substratum (e.g. amphipods, gastropods); and (3) organisms that live completely or partially buried, digging out small galleries in the substratum (e.g. bivalves, cumaceans, polychaetes). Although a diversity of prey was available to C. hakodatei, amphipods and mysids were the dominant food items with respect to area, season, size class, and sex. Amphipods are dominant in diverse habitats and play an important role in the food chain of marine ecosystems as the main food source for other predatory animals (Crawford 1937; Zhang 1974). Mysids are a ubiquitous component of zooplankton assemblages in a variety of aquatic environments and at times can comprise a significant proportion (up to 20%) of the total zooplankton biomass (Mauchline 1980; Price 1982; Webb et al. 1987; Grange & Allanson 1995; Jerling & Wooldridge 1995; Froneman et al. 2000).

The occurrence of major crustacean items in the shrimp stomachs was not related to size, sex, or area. Mysids and amphipods were found consistently in each shrimp size class, with mysids being the dominant food item. It has been reported that in silt and clay areas, meiofauna and macrofauna are important biological components (Pihl & Rosenberg 1984), while sand areas are dominated by calanoid copepods (Boddeke et al. 1986). The results of our investigation are similar to those reported in other studies of Crangonidae shrimp. Members of this family are almost exclusively carnivorous, including C. affinis by Kosaka (1970) and Hong and Oh (1989), C. allmani by Allen (1960), C. crangon by Lloyd and Yonge (1947), Tiews (1970) and Oh et al. (2001), C. franciscorum and C. nigricauda by Wahle (1985), C. septemspinosa by Price (1962) and Wilcox and Jeffries (1974), and C. uritai by Nakaya et al. (2004).

The most important factor affecting the diet of Crangon spp. is the spatial and temporal availability of prey (Wilcox & Jeffries 1974; Wahle 1985; Oh et al. 2001), with amphipods being the most important benthopelagic prey animals (Venables 1981; Moller & Rosenbreg 1982). Spatial differences in the diet of C. hakodatei are related to differences among habitats, particularly with respect to substrate type, which would determine the abundance and population structure of potential prey types in the community. Based on many studies it appears that Crangon species adopt an opportunistic strategy, which suggests in turn that the diet of C. hakodatei may shift depending on habitat type and the availability of potential prey.

In this study, the substantial differences found for the stomach fullness of sand shrimp among areas could not be related to the availability of food items, because the dominant and preferred trophic groups of macrobenthic organisms did not change markedly among seasons at either site. Stomach fullness is affected by both extrinsic (season) and intrinsic (sex) factors (Lloyd & Yonge 1947; Meredith 1952; Henderson & Holmes 1987). In this study, males had lower values for stomach fullness, food quality indices, and evenness than females. Both sexes consumed the same prey items, but showed different abundances and frequencies of occurrence of prey. These results indicate a slightly greater predatory ability for females. These differences could also be attributed to sexual dimorphism and to size differences between the sexes. There was no substantial positive relationship between C. hakodatei size and prey size. The trophic diversity indices varied little among areas but were higher for the larger size class. The highest values were observed in spring, probably reflecting a substantial increase in the availability of prey items in this season, while the lowest values occurred in autumn as a result of a reduction in the range of prey items. The results may also reflect changes in prey abundance associated with seasonal variations in primary production.

Additional information

Notes on contributors

Kyung-Jun Song

These authors are contributed equally to this work