Stomach contents from 809 king crabs, Paralithodes camtschatica (Tilesius), from 6 areas near Kodiak Island,. Alaska, and 9 sampfing periods (1978-1979) ...
MARINE BIOLOGY
Marine Biology 66, 243-250 (1982)
9 Springer-Verlag 1982
Food and Feeding Habits of the King Crab Paralithodes camtschatica Near Kodiak Island, Alaska * S. C. Jewett and H. M. Feder Institute of Marine Science, University of Alaska; 905 KoyukukAvenue North, Fairbanks, Alaska 99701, USA
Abstract
Stomach contents from 809 king crabs, Paralithodes camtschatica (Tilesius), from 6 areas near Kodiak Island, Alaska, and 9 sampfing periods (1978-1979) were examined quantitatively; 713 (88%) contained food. Mollusca (mainly the bivalves Nuculana spp., Nucula tenuis, and Macoma spp.) and Crustacea (mainly barnacles) were the dominant food groups in terms of percentage wet weight and frequency of occurrence; fishes were the next most important group of prey. No significant differences in feeding between sexes occurred; however, significant differences were apparent in the quantity of food consumed from different sampling periods, areas, depths, size groups, and crab molt-classes. Consumption was greater in spring and summer and in offshore locations at depths of 126 to 150 m. In addition, king crabs < 140 mm carapace length (CL) consumed more food than crabs >--140mm CL. Adult, newshell (individuals that molted during the last molting period) females greater than 95 mm CL, and newshell males greater than 100 mm CL, each contained more food than did juvenile, newshell females < 120 mm CL.
Introduction
The king crab Paralithodes camtschatica (Tilesius) is the most important commercial shellfish in Alaska. The king crab harvest around Kodiak Island from 1960-1961 through 1980-1981 was 266.9 X 10~ t (metric tons); during the 1980-1981 season, 9.3• 10s t of king crabs yielded nearly 23.5 million dollars to fishermen [G. C. Powell, Alaska Department of Fish and Game (ADF&G), personal communication, 1981]. A major prerequisite for management of Alaska's king crab fishery is knowledge of the crab's distribution, abun* Contribution No. 449, Institute of Marine Science, University of Alaska, Fairbanks, Alaska 99701, USA
dance, and behavior. Investigations on the food habits and feeding behavior of king crabs have been mainly conducted on the west Kamchatka Shelf and the Okhotsk Sea (Marukawa, 1933; Feniuk, 1945; Logvinovich, 1945; Vinogradov, 1945; Kun and Mikulich, 1954; Kulichkova, 1955; Takeuchi, 1959, 1967; Tarverdieva, 1976); some studies have also been carried out in the Sea of Japan (Nakazawa, 1912; Kajita, 1925; Ishii et al., 1929; Kajita and Nakazawa, 1932; Marukawa, 1933), the southeastern Bering Sea (Kawasaki, 1955; McLaughlin and Hebard, 1961; Chebanov, 1965; Cunningham, 1969; Feder and Jewett, 1980), and Cook Inlet (Feder et al., 1980; Feder and Paul, 1981). Limited semi-quantitative feeding data are available for king crabs from two bays of Kodiak Island (Feder and Jewett, 1977). The present paper includes extensive quantitative feeding data on this species from additional inshore and various offshore regions of Kodiak Island.
Materials and Methods
Specimens ofParalithodes camtschatica (Tilesius) were obtained near Kodiak Island by trawl and SCUBA diving. Inshore areas sampled by trawl included Izhut Bay located on the southeast side of Afognak Island, and Kiliuda Bay located on the east side of Kodiak Island (Fig. 1). Inshore areas sampled by SCUBA were: Near Island Basin, McLinn Island, and Anton Larsen Bay. Nine stations at various locations outside the bays along the Kodiak Shelf were sampled by trawl. The depths of stations in Kiliuda Bay, Izhut Bay, and on the Kodiak Shelf were 0 to 100 m, 175 to 200 m, and 50 to 200 m, respectively; diving location depths were 0 to 25 m. Data were collected during 10 cruises in 1978-1979. The NOAA ship "Miller Freeman" was used for offshore sampling, and the M.V. "Yankee Clipper" and the R.V. "Commando" were used for inshore collecting. Sampling from the "Miller Freeman" was conducted from 19 June to 9 July 1978 and 14-24 February 1979 using a commercial0025-3162/82/0066/0243/$ 01.60
244
S.C. J ewett and H. M. Feder: Food and Feeding of King Crabs 155 o I
i ALASKA
153~
154 ~ I
....]
15l ~
152 ~
I
I
I
i 1
-I
z h u t Bay IS.
I
n t o n Larsen Bay Island Basin
] ~ i;
o
..:~ ..
K6di~k C ~ M c L i n n
Island
IS L A N
.
"
....
;~Y--Kil/uda Bay
7 ~
-- 57~
?
. . . . . 0
/ t56 ~
I 155~
I
I
154~
Fig. 1. L o c a t i o n s where k i n g crabs, P a r a l i t h o d e s camtschatica,
153~
50 kilometers
l 152 ~
100
--
56~
I 151~
were collected f o r stomach analyses, ] 9 7 8 - 1 9 7 9 . N u m b e r s represent s
stations in the deeper waters of the Kodiak Shelf area
size 400/~m mesh Eastern otter trawl (12.2 m horizontal opening). The "Yankee Clipper" sampled from 10 to 22 April, 7-15 May, 7-22 June, 9-21 July, and 8-23 August 1978. The " C o m m a n d o " sampled 7-15 May, 7-22 June, 9-21 July, 8-23 August, and 4-17 November 1978, and 1-20 March 1979. A try net (small otter trawl; with 6.1 m horizontal opening) was deployed from the "Yankee Clipper", and a try net and Eastern otter trawl were used from the "Commando". Diving via SCUBA was conducted in May and June 1978 and May 1979. Crabs were measured (carapace length in ram) and weighed (blotted wet weight in g). Carapace length (CL) is defined as the distance from the posterior margin of the right orbital indentation to the mid-point of the posterior marginal identation. Crabs were categorized in one of 8 classes (adapted from Powell et al., 1974): (Class 1) juvenile, newshell females < 120 mm CL - individuals that
molted during the last molting period; (2) adult, newshell females > 95 m m CL; (3) newshell males < 100 m m e L ; (4) oldshell males < 100 m m CL - individuals that failed to molt during the last molting period; (5) very oldshell males < 100 m m CL - individuals that failed to molt during the last two or more molting periods; (6) newshell males > 100 m m e L ; (7) oldshell males > 100 m m e L ; (8) very oldshell males > 100 m m eL. Stomachs (which included the posterior portion of the esophagus) were removed, placed in plastic bags, and fixed in 10% buffered formalin. In the laboratory, stomach contents were sorted by taxon. Prey items in each taxon were blotted dry, weighed to the nearest 0.001 g, and measured volumetrically by water displacement to the nearest 0.01 ml. Weights were obtained by weighing a vial with a known quantity of water and then weighing the vial and water, plus the prey items. The difference between the two weights represents the
S. C. Jewett and H. M. F eder: Food and Feeding of King Crabs taxon weight. The percentage wet weight of each food item was calculated by dividing the wet weight of a particular item in all crabs by the total wet weight of all food in all crabs. Prey organisms were also quantified as the percentage of stomachs containing various food items from the total number of feeding individuals. In this paper, all items in stomachs, including skeletal material, are reported as food, Items reported as bivalve mollusk prey consisted of approximately 50% shell material and 50% soft parts. Barnacle prey was mainly shell material. Fish remains were not identified to species. King crab stomach fullness was calculated using a method adapted from Cunningham (1969), who related carapace length and maximum stomach volume by the regression equation, Y= 34.25-0.72 x + 0.0047 x (r = 0,899). Crabs examined in our study were similar in size to those examined by Cunningham, thus we used his regression equation with our data to calculate maximum stomach volumes. The percentage of fullness was derived by dividing the observed volume of food contents by the maximum stomach volume for a particular size of crab. The Spearman rank correlation (Zar, 1974) revealed how well weight and volume (fullness) of food contents in crab stomachs correlated. A correlation coefficient of 0.89 (P = 0.001) showed that the weight and volume of stomach contents were strongly related. Since prey weight is a quantitative measure commonly used in other investigations (e.g. Kun and Mikulich, 1954; Kulichkova, 1955; Takeuchi, 1959, 1967; Tarverdieva, 1976), this measure was used in most statistical analyses in the present study. Volume was the measure chosen to examine food contents relative to crab size. The weight of stomach contents was not normally distributed; therefore, the median weight value, rather than the mean, was used as the measure of central tendency. This median weight value was used in the statistical analysis in feeding between sexes. Three statistical procedures were used in analyzing the food of crabs. The first procedure, the Wilcoxon signed-ranks test (Zar, 1974), is a nonparametric test designed to test difference between paired observations. This test was used in determining feeding differences between sexes. The second procedure, the Kruskal-Wallis test (Zar, 1974), is also a nonparametric test used to test the hypothesis that all treatment effects are the same versus the null hypothesis that not all treatment effects are equal. This test was used in determining feeding differences between sampling periods, areas, depths, crab sizes, and crab classes, using rank sum values of food weights. A P-value was calculated and, if found to be less than 0.05, a third procedure was employed to make multiple comparisons using the rank sums (Dunn, 1964).
Results
Food Stomach contents of 809 specimens of Paralithodes camtsehatica from 6 areas and 9 sampling periods were
245 examined; 713 stomachs (88%) contained food. There was no significant difference in the quantity (median weight) of food utilized by males and females (Wilcoxon rank sum test; Zar, 1974; P=0.469); thus, food data for both sexes were combined. Prey items belonged to 6 major food groups: Annelida (Polychaeta), Mollusca, Arthropoda (Crustacea), Echinodermata, Pisces, and plant material (Table 1). Mollusks and arthropods were the dominant food groups in terms of percentage wet weight and frequency of occurrence; fishes were the next most important group. Bivalves were the dominant mollusk subgroup, accounting for 31.3% of the total food wet weight. The clams Nuculana spp., Nucula
Table 1. Paralithodes camtschatica. Important food items in stomachs of 713 king crabs collected from Kodiak Island area, 1978 - 1979. na: percentage frequency of occurrence not available Food items
Percentage wet wt of food
Percentage frequency occurrence
Annelida (total) Pectinariidae
2.5 0.2 1.8 0.5 32.5 31.3 11.2 0.4 3.8 0.5 0.5 1.1 0.3 6.9 0.4 0.6 0.6 0.9 3.9 1.2
40.0 11.6 13.9 na 78.5 72.3 31.3 4.1 24.5 1.0 17.1 15.4 2.5 26.6 4.3 6.0 7.0 6.6 na 25.1
0.5 0.7 31.4 19.2 1.9 3.0 2.6 4.7 2.8 1.1 0.8 0.3
3.8 na 59.0 24.8 3.1 2.9 14.2 na 18.5 1.3 10.5 1.5
0.6 12.7 13.5 2.9 1.3 0.4
na 17.4 52.7 34.8 26.6 33.7
Oweniafusiformis
Other annelids Mollusca (total) Bivalvia (total) Nuculana spp. Yoldia spp. Nucula tenuis Mytilus edulis Axinopsida serricata Clinocardium spp. Serripes groenlandicus Macoma spp. Mya spp. Protothaca staminea Hiatella arctica Pandora sp.
Other bivalves Gastropoda (snails) (total) Trichotropis spp. Other gastropods Arthropoda (total) Balanus spp. Pandalus si0p. Pinnixa occidental& Chionoecetes bairdi
Other arthropods Echinodermata (total) Asteroidea Ophiuroidea Strongylocentrotus droebachiensis
Other echinoderms Pisces Unidentified animal material Plant material Unidentified material Sediment
246 tenuis, and Macoma spp. made up 21.9% of the food weight, or 70.0% o f the total bivalve weight. At least 9 other bivalve genera were identified as king crab prey. A m o n g the arthropods, barnacles were the most important food. Two species, BaIanus crenatus and B. hesperius, occurred in stomachs, but the fragmented nature of these barnacles precluded quantification of these species. B. crenatus is the dominant barnacle in the study area. Fishes, which accounted for 12.7% of the food weight, were consumed at all sampling locations and during most periods. Fishes were particularly important prey in Izhut Bay where they accounted for 68 and 92% of the total prey weight in June and July, respectively. The prey taxa taken within any region were usually diverse. Crabs from the shallow waters of Near Island Basin, McLinn Island, and Anton Larsen Bay contained 21 to 53 different prey taxa. A m o n g the 98 prey taxa taken by crabs from the deeper waters o f the Kodiak Shelf, 73 taxa were identified at a single station (Station 9) with 25 taxa identified from a single crab (also at Station 9). The n u m b e r of prey taxa in crabs was lowest in Izhut Bay. The mean amount of prey found in feeding individuals was 1.41 g or 0.001 g of prey per gram crab.
S.C. Jewett and H. M. Feder: Food and Feeding of King Crabs Table 2. Paralithodes camtschatica. Kruskal-Wallis one-way ANOVA for king crab food wt relative to sampling periods Sampling period
Apr. 1978 May 1978 June 1978 July 1978 Aug. 1978 Nov. 1978 Feb. 1979 Mar. 1979 May 1979
Stomachs with food No.
%
15 84 182 261 30 49 17 30 45
32 100 99 92 70 89 77 79 83
Average rank sum of contents
3 715.5 24 121.0 65 436.5 106 743.0 6 753.0 14 545.5 3 323.5 9 855.5 20 057.5
247.7 287.1 359.5 408.9 225.1 296.8 195.5 328.5 445.7
Pairs significantly different (P< 0.05) b Apr. 1978 < May 1979 May 1978 < July 1978 May 1978 < May 1979 June 1978 > Aug. 1978 June 1978 > Feb. 1979 July 1978 >Aug. 1978 July 1978 > Nov. 1978 July 1978 > Feb. 1979 Aug. 1978 < May 1979 Nov. 1978< May 1979 Feb. 1978 < May 1979
Sampling Period, Area and Depth King crabs were collected monthly from April-August 1978, N o v e m b e r 1978, and February, March, and May 1979. A significant difference was found in the amount of food in stomachs between months (Table 2). In general, crabs from spring-early summer months (May, June, and July) contained significantly more food than crabs from late summer-fall-winter months (August, November, and February). When crabs from the 6 sampling areas (Near Island Basin, McLinn Island, Anton Larsen Bay, Kiliuda Bay, Izhut Bay, and Kodiak Shelf) were compared, a significant difference in food weight was detected, i.e., crabs from the Kodiak Shelf stations contained significantly more food than crabs from McLinn Island and Kiliuda Bay. In order to determine which Kodiak Shelf stations were most important, multiple comparisons were made. The significantly different pairs that resulted were Station 9 > Station 7; 9> 8; 12>7; and 12> 8. Crabs from Stations 9 and 12 were subsequently combined, compared with crabs from the nearshore areas (Near Island Basin, McLinn Island, Anton Larsen Bay, Kiliuda Bay, and Izhut Bay), and found to contain significantly more food than crabs from all nearshore areas (Table 3). Crabs from all nearshore areas were similar. Depths at which king crabs were collected were divided into 8 strata; 1 6 0 - 179 120 - 139 > 140- 159 120 - 139 > 160 - 179
Calculated test statistic=52.66. Calculated P-values ~ 0 assuming a 2a distribution with 7 DF b Multiple comparison test (Dunn, 1964)
100 93 91 84 85 81 20
c o m p a r e d by various size groups, a significant difference was f o u n d : crabs < 140 m m C L h a d fuller s t o m a c h s than crabs > 140 m m C L ( T a b l e 6). M o s t crabs e x a m i n e d b e l o n g e d to one o f two classes: n e w s h e l l o v i g e r o u s females > 95 m m C L (Class 2) (37%), a n d n e w s h e l l m a l e s > 100 m m C L (Class 6) (30%). O l d shell males > 100 m m C L (Class 7) and n e w s h e l l j u v e n i l e f e m a l e s < 120 m m C L (Class 1) a c c o u n t e d for 17 a n d 11%, respectively. N e w s h e l l a n d oldshell males < 100 m m C L (Classes 3 and 4, respectively) a n d v e r y oldshell m a l e s
25> 76-100 25
