Food habits and consumption rates of common dolphinfish (Coryphaena hippurus) in the eastern Pacific Ocean.

Abstract--An ecosystem approach to fisheries management requires an understanding of the impact of predatory fishes on the underlying prey resources. Defining trophic connections and measuring rates of food consumption by apex predators lays the groundwork for gaining insight into the role of predators and commercial fisheries in influencing food web structure and ecosystem dynamics. We analyzed the stomach contents of 545 common dolphinfish (Coryphaena hippurus) sampled from 74 sets of tuna purse-seine vessels fishing in the eastern Pacific Ocean (EPO) over a 22-month period. Stomach fullness of these dolphinfish and digestion state of the prey indicated that diel feeding periodicity varied by area and may be related to the digestibility and energy content of the prey. Common dolphinfish in the EPO appear to feed at night, as well as during the daytime. We analyzed prey importance by weight, numbers, and frequency of occurrence for five regions of the EPO. Prey importance varied by area. Flyingfishes, epipelagic cephalopods, tetraodontiform fishes, several mesopelagic fishes, Auxis spp., and gempylid fishes predominated in the diet. Ratios of prey length to predator length ranged from 0.014 to 0.720. Consumption-rate estimates averaged 5.6% of body weight per day. Stratified by sex, area, and length class, daily rations ranged up to 9.6% for large males and up to 19.8% for small dolphinfish in the east area (0-15[degrees]N, 111[degrees]W-coastline). Because common dolphinfish exert substantial predation pressure on several important prey groups, we concluded that their feeding ecology provides important clues to the pelagic food web and ecosystem structure in the EPO.

**********

Dolphinfishes (Coryphaena hippurus and C. equiselis) are abundant, wideranging, epipelagic predators in tropical and subtropical oceans (Palko et al., 1982). They support important commercial, artisanal, and recreational fisheries in several regions (Beardsley, 1967; Oxenford and Hunte, 1986: Patterson and Martinez, 1991; Campos et al., 1993: Norton and Crooke, 1994: Lasso and Zapata, 1999). Dolphinfishes are also a large component of the bycatches of the tuna purse-seine and longline fisheries in the Pacific Ocean (Lawson, 1997; IATTC, 1999). They are commonly found near natural and artificial floating objects (Kojima, 1956: Hunter and Mitchell, 1966: Gooding and Magnuson, 1967; Wickham et al., 1973), a trait which facilitates their capture.

Calls have been issued for developing an ecological approach to fisheries management, taking greater note of species interactions and underlying ecosystem dynamics (FAO, 1995: Larkin, 1996: Mangel et al., 1996: Botsford et al., 1997). Removal of predator biomass by commercial fishing represents a "top-down" disturbance of the system. Selective exploitation of apex predators can have profound effects on pelagic ecosystems because of the removal of predation pressure (Essington et al., in press) and because of top-down, trophic-cascade effects (Shiomoto et al., 1997; Estes et al., 1998; Verheye and Richardson, 1998). An understanding of how top-down processes influence the dynamics of marine communities derives from a basic understanding of the trophic connections and rates of food consumption of the predators. Although four studies have provided limited data on the food habits of dolphinfishes in coastal areas of the eastern Pacific Ocean (EPO) (Hida, 1973; Campos et al., 1993; Aguilar-Palomino et al., 1998; Lasso and Zapata, 1999), little is known of the predation dynamics of dolphinfishes over the majority of their oceanic habitat.

Common dolphinfish (C. hippurus) are renowned for their rapid rates of growth and metabolism. In Hawaiian waters, common dolphinfish have attained average lengths of 120 cm and weights of 12.5 kg at 12 months of age (Uchiyama et al., 1986). Standard metabolic rates of common dolphinfish are comparable to those of yellowfin (Thunnus albacares) and skipjack (Katsuwonus pelamis) tunas (Benetti et al., 1995). All three species have large surface areas and thin blood-water interfaces in their gills, morphological features that permit high oxygen diffusion capacity and elevated metabolic rates (Brill, 1996). High energy requirements imply that predators like dolphinfish can account for important amounts of tertiary production removed from an ecosystem (Essington et al., in press), but rates of food consumption by dolphinfish in nature have not been measured.

The objectives of our study were to define the trophic relations of the common dolphinfish, and their prey in both coastal and oceanic areas of the EPO and to provide preliminary estimates of their daily rates of food consumption.

Materials and methods

The dolphinfish were caught by tuna purseseine vessels of Colombian, Mexican, Panamanian, and Venezuelan registry from December 1992 through September 1994. The fish were caught as bycatch of the purse-seine fishery for tunas associated with dolphins, with floating objects, and as unassociated schools ("schoolfish"). In dolphin sets, the net is deployed around aggregations of primarily yellowfin tuna and spotted or spinner dolphins (Stenella attenuata or S. longirostris) (or both dolphin species) after a high-speed chase by speedboats. Floating-object sets are made by encircling flotsam, commonly tree parts and artificial fish-aggregating devices (FADs), and associated fauna with the purse seine, usually in the early morning. Schoolfish are detected by seabird activity and disturbance of the water surface caused by the fish swimming just below. The species composition and size and age distribution of the fauna are distinctly different for the three aggregation types and fishing strategies (Hall, 1998).

Stomach samples

Common dolphinfish stomach samples were taken at sea by observers of the Inter-American Tropical Tuna Commission (IATTC). The purse-seine sets yielding the dolphinfish samples were distributed across the geographical range of the EPO tuna fishery at that time (Fig. 1). We obtained samples from 74 purse-seine sets over a 22-month period: 61 sets (82%) were made on floating objects; 4 sets were made on dolphins; and 9 sets were made on unassociated tuna. On board the vessels, the observers measured the fork length (mm) of each dolphinfish, determined the sex if possible, and excised and immediately froze the stomachs. In the laboratory, we thawed the stomachs and visually estimated the stomach fullness as a percentage of the stomach capacity. Then, we identified the stomach contents to the lowest taxon possible, weighed them to the nearest gram, and enumerated them when individuals were recognizable. The counts of paired structures, such as cephalopod mandibles and fish otoliths, were divided by two to estimate numbers of prey. We categorized the digestion state of the prey: 1 = intact or nearly intact; 2 = soft parts partially digested; 3 = whole or nearly whole skeletons without flesh (or comparable state for nonfish taxa); and 4 = only hard parts remaining (primarily fish otoliths and cephalopod mandibles). We measured the length, or maximum dimension of individual prey to the nearest mm, if sufficiently intact. For cephalopods, we recorded the mantle length excluding tentacles.

[FIGURE 1 OMITTED]

Identifying the prey depended on the digestion state of the remains. We used the following keys to identify fish prey in digestion state 1: Jordan and Evermann (1896), Meek and Hildebrand (1923), Parin (1961), Miller and Lea (1972), Thomson et al. (1979), Allen and Robertson (1994), and Fischer et al. (1995b and 1995c). When the fishes were digested to state 2 or 3 we used taxonomic keys of vertebral characteristics (e.g. Clothier, 1950; Monod, 1968; Miller and Jorgenson, 1973) and compared skeletons of whole fishes collected in the EPO. We identified the crustacean prey from exoskeleton remains using the keys of Garth and Stephenson (1966), Brusca (1980), and Fischer et al. (1995a). We identified cephalopod prey from mandible remains (Clarke, 1962; Iverson and Pinkas, 1971; Wolff, 1982; Clarke, 1986). The fish collections at Scripps Institution of Oceanography and the Natural History Museum of Los Angeles County, and the cephalopod collection at the Santa Barbara Museum of Natural History were used to compare and validate prey identifications.

Data analysis

We analyzed the diet data by calculating three diet indices for each prey taxon. We calculated gravimetric importance of the prey (%W) as percentages of the total prey weights, numerical importance (%N) as percentages of total counts, and frequency of occurrence as the number of dolphinfish stomachs that contained a particular prey. We calculated percent occurrence (%O) as a percentage of all the dolphinfish sampled, regardless of whether their stomachs contained food. We present these three indices by prey taxon in detailed tables, summarized at several levels of taxonomic resolution. To facilitate analysis, we also grouped the prey taxa by order (e.g. Tetraodontiformes), family (e.g. Carangidae), genus (e.g. Auxis spp.), or functional group (e.g. flyingfishes).

Because the three diet indices provide different insights into predation habits, we applied a graphical representation of these measures, proposed by Cortes (1997), to help interpret the data. We made three-dimensional scatter plots of %O, %W, and %N for all samples and for the data pooled by sampling area to help evaluate the degree of dominance of particular prey and the feeding strategy (generalized vs. specialized) of the dolphinfish. Although we measured the three components of the index of relative importance (IRI) (Pinkas et al., 1971), we did not calculate IRI values because the index is dependent upon the taxonomic resolution of the prey (Hansson, 1998). Also, for a predator that consumes a large size range of prey (see heading "Prey size," below), the IRI is overly influenced by numerous small prey.

We examined diel feeding characteristics by stratifying the data according to stomach fullness of the predator and digestion state of the prey. The scheme for grouping the data, patterned after Calliet (1976), is diagrammed in Figure 2. Prey in digestion states 1 and 2 were categorized as from "recent" feeding events, whereas prey in states 3 and 4 were categorized as from "previous" feedings. These two strata were further subdivided according to stomach fullness. Prey from stomachs [less than or equal to] 50% full were categorized as "low" fullness or empty, whereas those from stomachs >50% full were categorized as "high" fullness (Fig. 2). We plotted the percent occurrence of the prey items in these four digestion and fullness strata by area and the time of day the sets were made: "early morning" (05:12-09:00),"late morning" (09:01-12:00), "early afternoon" (12:01-15:00), and "late afternoon" (15:01-18:16 hours).

[FIGURE 2 OMITTED]

We fitted regression trees (Breiman et al., 1984) to the gravimetric data for each prey group to detect statistically important differences by area and dolphinfish size. Regression trees are well suited for detecting and extracting important relations and complex interactions in multivariate ecological (De'ath and Fabricius, 2000) and fisheries data (Watters and Deriso, 2000). We used a two-step process. For both steps, the %W of each prey group in the stomach contents was the response variable. For the first step, defining area strata (see next paragraph), we used latitude and longitude as the predictor variables. For the second step, modeling the importance of area and dolphinfish size in explaining variation in the %W for each prey group, we used area designations (north, west, east, southwest, and southeast) and fork length as the predictor variables. We used the tree functions in S-Plus (MathSoft Inc., 1999) and cross-validation to prune fully grown trees so that only important splits remained. Prediction errors were used as pruning criteria (Breiman et al., 1984; De'ath and Fabricius, 2000).

We stratified the data by area (Fig. 1) according to two criteria. Latitude divisions at 15[degrees]N and 0[degrees] were based on the spatial and seasonal heterogeneity of the purse-seine sets that provided the samples. All the sets sampled from May through November each year were made north of the equator, and most sets sampled during December through April were made south of the equator. Also, the regression trees indicated that latitude and longitude were important in explaining the variability in the gravimetric data for several prey taxa. Epipelagic-cephalopod taxa were most important in the diet of the common dolphinfish caught east of 82[degrees]41'W and south of 1[degrees] 46'S, and flyingfishes were most important in the diet of those caught west of 81[degrees]W. Therefore, we stratified the data from south of the equator into "southwest" and "southeast" areas separated at 82[degrees]41'W (Fig. 1). Similarly, we stratified the data from samples collected between 0[degrees] and 15[degrees]N into "west" and "east" areas divided at 111[degrees]W because a regression tree fitted to the gravimetric data for the Tetraodontiformes indicated that this meridian was important in explaining variation in %W for this taxon. Tetraodontiformes were more important in the diet of the fish caught west of 111[degrees]W.

Consumption rates

We employed a method described by Olson and Mullen (1986) to calculate preliminary estimates of daily rates of food consumption by common dolphinfish. The model predicts feeding rate (r, grams per hour) by dividing the mean weight of the stomach contents per predator ([bar]W, grams) by the integral (A, proportion x hours=hours) of the function that best fits experimental gastric evacuation data. For a predator that consumes a variety of prey that are evacuated at different rates,

(1) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII],

where subscripts i refer to each of I prey types. Daily meal is r multiplied by 24 h for fish that feed both day and night. Daily ration is daily meal expressed as a percent of body weight. We estimated the body weight of each dolphinfish from the length, according to the relationships of Lasso and Zapata (1999):

(2) M = a[L.sup.b],

where M = body weight (g); and L = fork length (cm).

They estimated that a = 0.0406, 0.0420, and 0.0224, and b = 2.6588, 2.6328, and 2.78 for males, females, and common dolphinfish of undetermined sex, respectively. We estimated daily consumption rates for dolphinfish …