Blbllat· International Council for the Exploration of the Sea 'J. , Fi'r:hlt.I, Ilar:t uu '" CM 1995/1:6 Baltic Fish Committee Ref.: C,E Consumption of food and evactiation in dab (Limandda Iimanda) related to saturation and temperature. Preliminary results. by Bagge, ale, Else Nielsen and lohn F. Steffensen • Abstract: Dab Oimanda limanda) accÜmatized to 4 different levels of oxygen saturation have been fed to satiety in intervals of 24 hours with food of known wdght. The food given was cooked Mytilus edulis and raw flesh of cod. The levels of oxygen saturation were 20-40 %, 35-50 %, 65-85 % and 80-90 %, the . temperatures were 11.5-12.0°C, 1l.Q-l1.5°C, 10.0-1O.5°C respectively and the salinity 30 %. The 3 lower levels of saturation were obtained by addition of nitrogen adjusted by ci computer program connected to 3 wrw OXY 196 arid 3 valves. • At the saturation level of 80-90 % the mean consumption per 100 g consumer/day or 04.5 % of the mean body weightlday. The amount of food eaten a given day seems to be related to the amourit eaten the day before. At the lowest level of saturation (20-40 %) the consumption may stop for several days in between varying between 0 and 3.0 gllOO g consumer/day or 0.3.0 % niean body weight. The mean consumption per 100 g consumer/day. at 80-98 % saturation in 94 consecutive days waS 4.5 times higher.thari at a saturation of 20-40 %• . Afte'r acclimatization untill normal feeding at two different temperatures 7.5-8.00C and . 10.5-10.6°C the dab have been starved for 24 hours and then offered pieces of cod flesh of known weight each with a numbered internal plastic tage .The food was offered to satiety. The oxygen saturation was 90-100 % and the salinity 33 %. . , . , With intervals of 6 hours (6-30 hours) 3-9 fish were killed. The content of food in the stomach and the first part of the intestine and in the last part of the intestine have beeri removed and weighed separately, and the initial weight estimated from the numbered tags~~e rates of evacuation the first 18 hours at the two different temperatures were at 1 1; Introdllction: In the transition area betZveen the Nofth Sea and the Baltic, the Kattegat and the Belt Sea, hypoxia regularily occurs in summer and autumn due to a distinct halocline between the brackish Baltic surface water and the saline North Sea water, along the bottom which do not allow oxygen from the atmosphere to supply the bottom water. These oxygen conditions mainly influence bottom invertebrates and demerSaI fish like plaice and dab. . . ' . Iri dab a reduction in growth has been indicated in th6 southem Kaitegat, Bagge and Nielseri, 1987. Bagge et aI., 1993 and in Arhus Bay, Bagge et al 1994. It may be assumed that hypoxia reduces .the appetite of the fish, which may be conriected with a redueed evaeuation of the stomaeh. Gwyther arid Grove (1981) .confinned for dab that stomaeh fulness is a major faetor controlling appetite in dab•.That is shown earlier for other species by several authors, Grove et al 1978, Elliot and Persson; 1978 and Brett (1971). This report is very preliminary. 2; l\faterial and Methods. e· The experiments on gastiic evacuatiori (A.) have been carried out in the laboratory of Danish Institute of Fishery and Marine Research in Charlottenlund and the experiments food optake reIated to oxygen saturation (B) has beeit carried out in the Marine BioIogical Laboratory (University of Copenhagen) in HeIsiilg0T. A. GaStric evacllation. Dabs were triiwIed in 10 minutes hauIs Oetober 1994 and January 1991 in the northem .' Sound (Subdivision 23) and kept in three 700 I-tanks with recircuIated aerated Sea water, satinity 33 0/00. The temperature was in the first experiment 7.5°C and in the second 10.5°C. The illumination during day lasted 10 hours in both exPeriments. 30 individuals were pul in each tank in the first experiment but after 1 month reduced by mortality to a total of 25 whieh were feeding regularilyon raw eod meat. In the seeond experiment only 15 individuals were put in each tank which yieIded 30 feedirig individuaIs after 2 weekS. In o~de.r ~o estlmate the individuaI food intake the fish were offererl lumps of eod meat of 2 known weight tagged with internal numbered yellow plastic tags 3 x 5 mm. These tags were easy to recognize in the stomaehs and the intestines and not regurgitated as in cod, possibly because in dab the stornach is not that distinctly separated from the intestine. .Before the experiment on stornach evacuation at 1O,5°C was started with tagged food the dab in the 3 tanks was fed with cod meat of known weight to estimate the mein food uptake per day relative to the mean body weight. The experiment was run at 1O,5°C to make it comparable with the experiment (B) in Helsingor run at that temperature. The following procedure was used to estimate the evacuation over time: the fish were starved 24 hours to secure empty stomaehs and then fed until satiation with tagged food of known weight. After 6, 12, 18 and 24 hours respectively 3-9 dab (table 2) were captured, weighted and measured individually and then killed. The content of stomaehs arid guts were then weighted separately and the original intake of food estimated from the numbered tags. • 3. Results. 3.1 The daily consumption: In the experiment run at 1O,5°C. the dab have been offered food until it was sure that all of them had started feeding (about 4 weeks). The fish were then fed until satiation 6 days per week with food of known weight the last 14 days before the evacuation experiment started in order to estimate the daily consumption relative to body weight. The results are shown in table 1 as weight of food relative to body weight in percent. It appears from the table that the food intake varies from 0,64 to 3.97 % of b. w. (body weight) and that a high food intake is followed by a day with a lower one. The values are means of 18,10 and 10 specimens in tank 1,2 and 3 respectively. These day to day fluctuations are also observed by Pandian (1970) ~d in the parallel experiments in Helsingor (B). 3.2. The gastric emptying over time. I • The gastric evacuation at 7.5°C and 1O.5°C as percentage of the feod intake after 6, 12, 18, 24 and 30 hours is shown in figure 1 and table 2 together with the means and the standard deviations of the weights of stornach content relative to body weight (b.w.) as :percentage. The means and standard deviations of the body weights and the standard deviations are given as weIl. Fitting a straight line to the percentual evacuation of stomaehs over time at the two temperatures it appears from figure 1 that at 7.5°C the emptying time is about 49 hours (48.9) and at 10.5 C. about 30 hours (30.4). Gwyther and Grove (1981) have given a none linear multiple regression which describes the relations between the gastric emptying time (GET hours), temperature, body weight and meal size relative to body weight: In(GET)=1.46+0.68(In M)+ 0.39 (In W)-0.035T where M is the relative meal size 3 . . ln(GET)= 1.46+0.68(In M)+ 0.39 (In W)-0.035T where M is the relative meal size (M%b.w.), W the body weight (W grams) and T the temperature. Fitting the data in table 2 to that model with T=7.5°C~ M=L73 and W=1.68.9 (mean of the five mean body weights in table 2) and T=10.5~C, M=2.76, W=207.5 (mean of the four mean be 35.5 hours and GET of 1O.5°C 47.6. body weights in tiible 2) the GET of 7.5°C The reason is that the impact of temperature inGwyther arid Grove's equation is negligible while the riieal size relative to body weight is important.. will The data from table 2 have been fitted to the none linear multiple regression model as usect by Gwyther anrl Grove (1981), (REG in SAS) and aR selection used to control the relevance of all the variables: Variables In(M) .. T In(W) In(M)T In(M)ln(W) In(W)T In(M)ln(W)T R2 0.2242 0.0773 0.0110 0.2792 0.2263 0.1099 0.2799 The regression is mainly influenced by In(M) and T. The weight of the fish is of negligible importance. Omission of In(W) yields the following expression: In(GET)~i65+0.41In(M)-0.087T (R2=0.2792, n=50) GET at 7.5°C. and 10.5°C. is then 25.1 hours and 23.5 hours respectively. B. Feeding :md Oxygen Saturation. 4. Material and Methods. - The experiments at the Marine Biological Laboratory (University of Copenhagen) in Helsingcr were started in the beginning of December 1994. The dab used in the . experiments were obtained as described in (A.). four. tanks height 45 cm, wicith 49 cm ancl length 100 cm were usect, covered on the : bottom with 5 cm cleaned sand and on the surface with a plate of PVC foam. The volume of water in 3 of the tanks (AI-A J ) was 100 I and in the fourth (~) 1701. Tanks A2 and AJ were corinected so that the water was renewed from a recirculated system in Ai and retumed to the system from AJ in order to create 3 different oxygen saturations. The wa.ter in Ao was renewed from the recirculatect system and retumect direct1y to the system. . 4 dab (21-26 cm) (95-175 g) were put in each of the tanks AI-A J and 5 in Ao• The . SaIinity wa.s 30 0/00 and the temperature in Al was 10.0-11.0°C increasing to 11.0-12.4OC in A3• In Ao the temperature was 9.2-9Jrc. 4 Al-A J and the temperature and oxygen saturations were continously down loaded in file format. 3 different oxygen saturations were obtained by covering the surface of the water in all tanks by plates of PVC foam and by admission of nitrogen to A2 adjusted by a computer program to keep the saturation at a level of 42 %. In that way a saturation of 35-50% was obtained in A2 and a saturation of 20-35 % in AJ • The temperature and oxygen saturation was also registrated manually every moming. In Ao only manual registration has been done. The fish were acclimatized in the period L 12.94-26.01.95 to eat at saturations: 80-90% in Ao• 20-35% in AJ , 35-50% in A2 and 65-85% in All being offered food of known weight (boiled MytHus edulis) every moming at 8 o'clock GMT. The registration of the daily feeding started 27.01.95. 5. results. 5.1 Saturations and temperature. The daily level of oxygen saturations in Ao (the highest level) and AJ (the lowest level) is shown in fig. 2 and the levels in between Al and A2 in fig. 3. It appears that there are fluctuations but that the lowest level AJ and the two highest levels Ao and Al are weIl separated. Tbe continous registration of saturation and temperature is not given but the mean saturations of the daily manual observations (940bs. in the period 27.01-5.02.95 are given below: Ao Al A2 AJ Mean saturation Std. Temp. 85,78% 72.69% 41.48% 29.80% 3.64 12.53 7.50 7.81 9.2-9.8 10.0-11.0 10.5-11.8 11.0-12.4 The standard deviation in Al is high but the levels of saturation in A2 and AJ differ significantlyas weIl from Ao and Al' 5.2 Tbe daily consumption. The daily consumption according to oxygen saturation is given as percent of body weight (b.w.). It was not possible to registrate the amount of food eaten individually, which means that a mean body weights per tank (Ao -AJ) have been used. The daily consumption is then: mean weight consumed per specimen Imean weight of consumers. 5 The mean weight consumed being the total amount consumed devided by the number of consumers in the tank irrespective whether all have eaten. During the experimental period growth of the dab was observed and therefore the mean weight of the consumers was changed 3 times during the period by mean of a length-weight key. In fig. 4a is shown the daily consumption per 100 g consumer in relation to oxygen saturation 20-40% and 80-90% and in fig. 4b in relation to oxygen saturations 35-50% and 65-85%.· At the 20-40% saturation manY days (45) without feeding are observed between days with food uptake, but also at ~~-90% saturation some few none feeding days occur (4). At 35-50% and 65-85% saturations the fluctuations are observed with a higher zero observations in the lower level of saturation (21) being compared to saturation 20-40% the number'of zero observations is reduced. • The day to day range of feeding rate and the mean feeding rate varies according to the oxygen saturation as follows. Ao Al A2 A3 Saturation % Mean Feeding rate % b.w. Mean St.d. CV 80-90 65-85 35-50 20-40 85.8 72.7 41.5 29.8 0-7.5 0-6.5 0-4.5 0-3.0 2~23 1.64 1.17 1.12 0.52 73.5% 73.5% 100.0% 106.0% 1.59 1.12 0.49 zero obs. 4 5 21 44 N 94 94 94 94 It appears that the maximum feeding rate and the mean feeding rate is decreasing with decreasing oxygen saturation and that ~e c.v. is increasing together with the increasing number of zero observations. The mean feeding rate of 80-90% and 20-40% differ significantly. , The' day to day fluctuations in feeding rates are as described in 3.2. Days with high . feeding rates are followed by days with low ones (pandian, 1981). 5.3 Daily oxygen saturation and consumption. Even if it was aimed to keep the 4 levels of oxygen saturation constant quite large fluctuations occurred except in the level 80-90%. The daily consumption expressed as % of body weight is compared with daily observations of oxygen saturation and correlated. 6 The variations are shown on fig. 5 a (80-90%), fig. 5b (65-85%), fig. 5 c (35-50%) and fig. 5 d (20-40%), and the resuIts of the correlations are given below: Ko1 K K2 K3 Saturation %Mean R N (N-2) P 80-90 65-85 35-50 20-40 85.8 72.7 41.5 29.8 0.0091 0.302 0.489 0.429 94 94 94 94 92 92 92 92 0.01 0.001 0.001 The strongest correlations are found for K2 and K3 which may indicate that below a saturation of 40-50% the appetite starts to be affected significantly. 5.4 The feeding rates in nature. During the years 1987-91 5928 stomaehs of dab have been sampled in the southem Kattegat on 4 stations, (Bagge et al. 1993) in March, May, September and November. • The mean stornach content as percent of body weight according to length in half cm for March, May and November including empty stomaehs 1987-91 is shown in fig. 6. To compare with the data from the experiments the feeding rates correspond to the lengths from 45-55 sem should be chosen (120-220 g. b.w.). It appears that the mean feeding rates in that interval are between 0.2 and 0.6% corresponding to a mean rate of 2.2 in the experiment at 80-90% saturation or 4-10 times less. It should be mentioned that the temperature in the experiment at 80-90% saturation was 9.2-1O.0CC compared to the mean temperature of the observations at the bottom (30 m) on the four stations in the southem Kattegat in March and May 1987-91 which was 5.85°C. However, the corresponding mean temperature in November was 12,4°C, and the stornach content relative to body weight does not differ in that month. (Fig. 7). : 7 • ----------- -------1 References. Bagge, O. and E. Nielsen, 1987. Growth and:recruitment of plaice in the Kattegat. ICES C.M. 1987/6:7, Demersal Fish Committee. Bagge, et al., 1993. Bundfaunaens betydning for bundlevende fisk i det sydlige Kattegat. Havforskning :fra Milj0styrelsen Nr. 27, 1993. • Bagge. 0., E. Steffensen, E. Nielsen and C. Jensen, 1994. Growth and abundance of plaice in Arhus Bay in relation to oxygen conditions 1959-1993. ICES, C.M. 1994/J: 16. Gwyther, D. and J. Grove, 1981. Gastric emptying in Limanda limanda (L.) and the return of appetite. J. Fish. Bio!. (1981) 18. 245-259. Pandian, T.J., 1970. Intake and conversion of food in the fish Limanda limanda exposed to different temperatures. Marine Biology 5, 1-17 (1970). 8 Stomach evacuation in dab. 110 100 90 80 ~ " g ;;; ... '. 60 -5 .E so .::: ..!! ~ • 70 .". ... 'q 40 30 ' ' '. '. '. '. '. '. 20 '. ". '. 10 '. '. '. 00 00 0 6 12 18 24 30 Hours. __ Evacuation at 7.fJC. Figure 1. 36 42 __ Evacuation at 10ft Evaluation of stomaehs at 7.5°C and 10.5°C. 0" Highest and lowest oxygen saturations. :7.01-:3.0~ 100 90 I SO ~ ." ~;; ~ <:5 70 B '" ~ 60 50 fvwG .10 30 20 '-'-------------------= 10 , --l 27.1 29.1 31.1 2.2 -l.2 6.2 S.2 10': l~ 14.: 16.2 18.2 20.2 22.2 24.2 16.2 28.1 23.1 30.1 1.2 3.2 5.2 7.2 9.2 11.2 13.2 15.2 17.2 19.2 21.2 13.2 25.2 27.2 Dal<. in 1995. ___ .-\3: saturation 20-JO%-. AO: saturation 80-90%. Highest and lowest oxygen saturations. 1.03-31.03. 100 90 SO i 70 ." i ;; 60 50 ". ~ C .10 20 10 L- '--_ _-'- -'--'--....J 1.03 3.03 5.03 7.03 9.03 11 13 15 17 19 21 13 25 27.3 29 31 2.03 4.03 6.03 8.03 10 12 14.3 16 18 20 22 :4 26 28 30 Date. in 1995. __ ,1.3: saturation 20-.100:'<. __ AO: saturation SO-90%. Highest and lowest oxygen saturations. 1.04-2.05. 110 100 90 • *- SO a ~ 60 Ö 70 50 .10 30 20 10 l....:.....:...'"-_-'-'-- --'- -'- '"--'-_-J 1.04 3.04 5.04 7.04 9.04 11 13 15 17 19 21 23 25 27 29 1.05 2.04 4.04 6.04 8.04 10 12 14 16 18 20 n 24.l 26 28 30 2.05 Date. in 1995. ___ A3: saruration 20-40%. Fig. 2. -+- ,.1.0: sarunuion SO-90~. Highest and lowest level of oxygen saturation. ~ '" ~ Oxygen saturations between highest an lowest. 27.01- 23.02 100 r--------------------------. 90 -i so 70 ~ .;;; 60 ~ so Ö JO 30 L...;.....:....-'--'-_-'--:.._-'-' 27 28 29 '--- ~ 31 2.02 a2 6.02 8.02 10 l! IJ 16 18 20 !2 2J 26 23 30 1.02 3.02 5.02 7.02 9.02 11 13 15 17 19 21 23 !5 27 Da,•• in 1995.Da,•• in 1995. __ AI:sa,uration 65-85%. __ A2:sa,uration 35-50%. OX'l'V~n satur3ticr:s ::e:'Ne~:"I ~ig!":!s: 3r:d lowest. 1.03-31.3 1~9;. 90 80 ~ .: 70 60 ;: ~ ~ 0 50 JO 30 20 1.03 3.03 5.03 7.03 9.03 ~.03 .l.03 Ö.03 3.03 11 10 13 12 15 U.3 17 16 19 13 21 :0 23 22 25 14 27.3 29 31 13 :;0 :6 Date. in 1995. ___ A1: ~tur:ltion 65 -S5~. -+- ..1.:: satur.1tion 35 - 50~c. Oxygensaturations between highes an lowest. I.OJ-2.05 !995. 100 . . . . - - - - - - - - - - - - - - - - - - - - - - - , 90 80 -i. .~ ~ a :; ~ C • 70 60 50 Ja 30 20 10 IOJ 3 OJ 51)4 7.04 9.04 11 13 15 17 19 21 23 !5 27 29 1.05 . 2.OJ· J.o.i 6.04 8.04 10 12 IJ 16 18 20 n 2U 26 23 30 2.05 Da'e< in 1995. ___ .~: saturation 3S-S0~. Fig. 3. -+- Al: S:ltur3tion 65-3S~. Levels between highest and lowest oxygen saturation. Consumption per 100 9 and oxygen saturation. D3b. \ ,,; 7 I 6 ~ '" ~ 5 ~ ~ ~ \ ~:; .~ z ~ 1 C 0 27.1 29.1 31.1 2.2 .1.1 6.2 8.1 10.2 12.2 1.1.2 16.2 IS.2 20.2 22.2 ~~.~ ~6.2 23.: 28.1 30.1 1.2 3.2 S.: ;.2 9.2 11.~ 13.1 15.2 li.2 19.1 .21.2 2.3.: 25.1 :i.2 Dates in 1995. .--..-\3 consumption at • ::O-JQ~. _ AO consumptian 3t 3O-90fi. Consumption per 100 9 and oxygen saturation. D.b. 3.5 . . . - - - - - - - - - - - - - - - - - - - - - - - . . . . , .. ~ :; ~ 2.5 . ; 1.5 e 1'" 1 § a1! 0.5 1.03 3.03 5.03 7.03 9.03 11 13 lS 17 19 21 23 2S 27.3 29 31 2.03 4.03 6.03 8.03 10 12 14.3 16 18 20 22 24 26 28 30 1.03-31.03 1995. __ A3: consumption at 20--lO %. -+- AO: consumption at 80-90 %. Consumption per 100 9 and oxygen saturation. • D.b. 3..$ . . . - - - - - - - - - - - - - - - - - - - - - - - . . . , . § ; ,j . ~~ ~ ~ 1.5 5, .i 1 f a1! 0.5 ~ 0'-- .........:....._.4'----...J ---........ ..--~'__----_+_-- 1.04 3.04 5.04 7.lJ..I 9.lJ..I 11 13 lS 17 19 21 23 2S 27 29 1.05 2.lJ..I 4.04 6.04 8.04 10 1: 14 16 18 20 22 24.l 26 28 30 2.05 D.'es in 1995. -+- A3: consumption at :O-.lQ~. Fig. 4 a. -+- AO: consumption 3t 80-90%. Consumption. relative to body weight at 80-90% and 20-40% oxygen saturation. Consumption per 100 9 and oxygen saturation. Dab. 6r---------------------..., 27.1 29.1 31.1 2.2 ~.2 6.2 8.2 10.2 12.2 14.2 16.2 18.2 20.2 22.2 2~.2 28.1 30.1 1.2 3.2 5.1 7.2 9.2 11.2 13.1 15.2 17.2 19.1 21.2 23.2 2.>.2 Dacos in 1995. ~ Al:c:onsumption at 65-85%. Al: consumption 3t 35-50%. -+- Consumption per 100 9 and oxygen saturation. Dab. '___ OL..--......~'__ _+_----__+'----l____J 1.03 3.03 5.Q3 7.03 9.03 11 13 15 17 19 21 2.03 ~.03 6.03 8.Q3 10 12 P.3 16 18 20 Dates in 1995. _ ....2: roruumption at 3$-50'-. _ z: 23 25 H 27.3 29 31 26 23 30 .... 1: consumption at 6$-85%. Consumption per 100 9 and oxygen saturation. Dab. 3,5 .....- - - - - - - - - - - - - - - - - - - - - - - - , • I • .i § 0,5 8 oL,.;..--l~----'_+_~-'---_++_:--=--=-"'__*--'-I___+_-_4I_----J 1.~ 3.~ 5.04 7.~ 9.~ 11 13 15 17 19 21 23 2S 27 29 1.05 2.~ 4.~ 6.~ 8.04 10 12 I~ 16 18 20 2! 2~ 26 !S 30 2.05 Daces in 1995. ___ Al:consumption at 65-85;0. Fig. 4 b. -+- A1::onsumprion <Je 35-S0~. Consumption relative to body weight at 35-50% and 65-85% saturation. Oxygen saturations and consumption. :7.U! -::'.0:. wo r - - - - - - - - - - - - - - - - - - - - - - , ~ 70 .= 60 50 20 10 27.129.1 ;1.1 2.2 .1..: 6.2 8.2 10.2 l:.~ l~.: t6.2 IS.: ~O.! ::.1 Z:.! 16.2 13.2 13.1 30.1 l.~ ;.~ 5.~ 7.:: 9.2 ! 1.113.: IS.: 17.: 19.: ::.:: ::3.1 ::5.: ~i.! Date; in 199;. ~ Consumpt!an pe::- 1000 ~ C:1nsumc::::. Oxygen saturations and consumption. 1.03-;1.03. 100 • 90 SO 70 ~ .= 60 SO i:l ;; .l() Ci 30 ~ 20 10 0 1.03 3.03 3.03 7.03 9.03 11 13 13 17 19 21 23 25 27.3 29 31 2.03 ~.03 6.03 3.03 10 12 P.3 16 18 20 22 24 26 28 30 Oalesin 1995. ~ Consumption per 1000 ~ consurcer. Oxygen saturations and consumption. 1.~-!.05. 110 100 • 90 SO ~ 'ia 70 60 30 ;; ~ C .10 30 20 10 0 1.04 3.04 5.04 7.~ 9.04 11 13 1S 17 19 21 13 25 27 29 1.05 2.~ 4.04 6.1)4 8.04 10 12 1~ 16 18 20 22 24.1 26 28 30 :!oO; Dates in 1995. ~ Consumption per 1000 ~ consumer. 5 a. Daily consumption and daily oxygen saturation (80-90%). Oxygen saturations and consumption. :~.nl-:s.~:. 110 100 90 8ll ~~ 70 .: 60 ~ ~ J 50 '" Ja 30 .- :0 10 0 27.1 ~9.1 31.1 1.2 ~.1 6.: 3.1 tO.~ l:.: 1.1.1 :6.: IS.: :0.: :1.: :J.: 1ö.: :3.: 23.1 30.! l.~ j.~ 5.~ i ..! 9...! 11.2 13.2 15.: 1';.: 19.121.: 13.: 25.: ::-.: Dut.:s in 1995. ~ C.;:)nsumpt:c.m pe:: 1000 'J ':;Jnsumc::-. Oxygen saturations and consumption. 1.03-3\.11•. 911 r----------------------, • 70 ~ 60 .: C 1.0.3.0. 5.0. 7.03 9.0. 11 G I! :7 19 11 13 25 27.• 29 .1 :m ~m 6.0. 8.0. 10 11 :~.. 16 18 20 1: 1~ 16 28 30 D~H~S in :9'15. ~ Consumpcion per 100U ~ l:onsumc::. Oxygen saturations and consumption. \.0~-2.l15. 1110 'Xl SO ~ • 70 tiO ~ ~ f 50 -lO .0 10 lil 0 I o~ •.().I 5.().I 7.().I 9.04 11 G 15 17 19 11 13 !5 27 29 \.Q5 . 1.0~ ~.().I 6.().I 8.04 10 12 I': 16 18 10 :: 1~.1 16 18 .0 2.05 Dates in 1995, ~ Consumpt:on pcr 1000 'l .:onsumc-:'. Fig. 5 b. Daily consumption and daily oxygen saturation (65-85%). Oxygen saturations and consumption. :-.Ol-::'.U:. 7U r----------------------, 60 '. '. x'" 10 o Oxygen saturations and consumption. 1.03-; 1.03. 60~--------------------, so • .~ ~ 30 ~ 10 1.03 3.03 5.03 7.03 9.03 11 13 15 17 19 21 2J 25 27J 29 31 2.03 4.03 6.03 8.03 10 12 l~J 16 18 20 ~ 2~ 26 28 30 Dates in 1995. ~ Consumpdon per 1000 g consumcr. Oxygen saturations and consumption. 1.~.-2.05. 60~----------------------, 50 • 10 1.04 3.04 5.04 7.04 9.04 11 13 15 17 19 21 2J 25 27 29 1.05 2.04 4.~ 6.~ 8.04 10 12 14 16 18 20 22 24.1 26 28 30 2.05 Oa,es in 1995. l88: Consurnp,ion per 1000 g consurncr. Fig. 5 c. Daily consumption and daily oxygen saturation (35-50%). Oxygen saturati?.n.s,~~d consumption. _ .11. _~.~_. 60...------------------_-----, 50 I .: ~ 30 '" .- 10 27.1 :!9.1 3l.1 2.2 .1.2 6.1 ~.~ HJ.~ 1:.2 lJ..~ 16.: 13.: ~O.2 2~.~ 2.1.1 :6.: 1S.': 2:tl 30.1 1.: 3.2 :.: i.2 ~.2 i !.: t3..= t5.': 17.: :9.:: :!.2 :3.1 ::S.~ :'7.: Da,", in 19~5. ~ Con~ump{ion ;')cr lOtJ0'l- -.:onsurnc:. Oxygensaturations and consumption. 1.03-31.0::. 60,-------------------------, • Q ';0 '" ,- 1.03 j.()j S.03 7.03 9.03 11 D 15 17 19 21 23 25 :7.3 29 31 ;'03 ~.Q3 6.03 S.03 10 11 1~.3 16 13 20 22 2~ 26 23 30 Da,es in 1995. ~ Consumption p<r 1000 g consum.::. Oxygensaturations and consumption. I.lIJ-::.U5. t\Il,------------------------, • 1.0~ 3.~ S.~ 7.~ 9.0~ .:.~ ~.~ 6.1J.l S.1J.l 10 11 13 1: I'; 17 19 21 :3 :!S 27 :9 1.05 16 IS :0 :: 1~.1 26 :3 30 :.05 D31es in 19'!5. 1~ ~ Co",ump'ion pe: 1000 ~ ~~nsumer. Fig. 5 d. Daily consumption and daily oxygen saturation (20-40%). Feeding rate relative to body weight per lengthgroup. Dab Kattegat. 0.8 0.7 • :§> ... 0.6 ~ >. ;( J!!. 0.5 = lJ .~ 0.4 ~ ~ 0.3 e"4 .=... ~ ... 0.2 :.> 0.1 0 16 21 26 31 36 41 46 Length groups in 0.5 cm. 51 56 61 __ 1987 -+- 1988 __ 1989 -e- 1990 -+- 1991 Fig. 6. Feeding rate relative to body weight per length group (%). Feeding rate relative to body weight per month and year. 0.8 , - - - - - - - - Dab Kattegat. --., 0.7 :0 .~ 0.6 >. ""j i Ei 0.5 " .~ ::: ~ 0.4 .Ff 0.3 ,.= " ""j 0.2 0.1 . . . . . . . . , ; , - - - - - - - - 1987 1988 1989 ....:.... 1990 --1 1991 Year. __ March. Fig. 7. __ ~1ay. __ 1989November. Feeding rate relative to body weight (%) per month and year. • .. • Day No. Tank 1 (No. 18) Tank 2 (No. 10) Tank 3 (No. 10) 1 2.21 1.54 1.07 2 1.60 .77 1.15 3 Not fed 2.46 .64 4 3.97 2.24 1.35 5 1.75 Not fed 1.20 6 2.06 Not fed Not fed 7 2.90 3.31 2.05 1.90 3.11 1.32 9 1.41 2.51 2.24 10 Not fed 2.03 2.12 11 2.80 2.55 .83 12 Not fed Not fed 1.06 13 Not fed 2.16 Not fed 14 Not fed 1.76 2.86 15 Not fed Not fed Not fed 16 Not fed 2.67 2.61 ~ble 1. The daily feeding relative to body weight at 10.5°C (Charlottenlund). - - ---- - - - - - - - - - - - - - - - - - - - - 1 . . Fish weight Stomach content (% b.w.) % left in stomach Hours Number Mean Std. Mean Std. 6 3 159.0 9.8995 1.730 .8282 100.00 12 5 172.3 31.3564 1.6679 .6160 76.56 18 7 186.6 68.8458 1.1215 .7832 75.81 24 4 148.3 40.5134 2.6730 1.4459 57.83 30 6 178.3 52.8711 0.7512 .42246 41.18 10.s°C. Fish weight Stornach content (% b.w.) % left in stomach Hours Number Mean Std. Mean Std. 6 9 191.2 41.7744 2.7265 0.7341 90.77 12 6 187.2 80.6874 1.732 1.0971 60.60 18 7 259.4 89.2185 1.812 1.6127 57.68 24 8 191.7 83.605 0.4167 0.3764 17.87 Table 2. Stomach content relative to body weight and time after feeding. Percentage of food left in the stomaeh.