Workshop Proceedings: Short Papers

Jarle Tryti Nordeide* and Jens Loss Finstad+
* Faculty of Fisheries and Natural Sciences, Bodø Regional University, N-8049 Bodø, Norway. + Master student at: Department of Fisheries and Marine Biology, University of Bergen, Norway


This paper reviews the first attempts to test the hypothesis that spawning calls of male migratory "Arctic" and a stationary "Coastal" cod group, which are sympatric during the spawning season, is a premating behavioural reproductive barrier. As predicted from the hypothesis, a hushed hubbub of sound with a transient character, band-width and harmonic spacing typical of cod calls, was revealed at a major spawning ground during the spawning period but not six months later. Moreover, individual calls from male cod kept in tanks varied a lot both in harmonic spacing and duration, from 42 to 79 Hz and 0.11 to 1.25 s, respectively. Such individual variation in calls is expected if females choose mate on the basis of their calls. However, the results so far has failed to support the third prediction, since no differences have been found between the calls of the two groups, neither in harmonic spacing, duration, or temporal structure of the calls.


North-east Atlantic cod consist of two stocks, the North-east Arctic, or "Arctic" cod, and the Norwegian coastal cod, or "Coastal" cod (Rollefsen 1934). Arctic cod migrates from the feeding areas in the Barents Sea to the spawning areas along the Norwegian coast and the most important spawning area is off the Lofoten Islands where the main spawning occurs in March and April (Bergstad & al. 1987). Coastal cod inhabit coastal areas and fjords, migrates short distances and spawns in a large number of fjords along the Norwegian coast (Rollefsen 1954), including off the Lofoten Islands (Hylen 1964). Both cod groups spawn in March and April and mature speciemens are sympatric during spawning at the major spawning grounds off the Lofoten Islands (Nordeide 1998). A controversial topic during decades has been whether or not the two cod groups interbreed, and a majority of studies conclude that they rarely do (see references in Nordeide & Pettersen 1998). If so, active partner choice is required, and lekking has recently been suggested to best describe the cod’s mating system (Hutchings & al. 1999; Nordeide & Folstad 2000).

Møller (1968) suggested that active partner choice based on acoustic calls may be an behavioural mechanism which prevents interbreeding between the two cod groups. The aim of this paper is to summarize the first attempts to test predictions derived from Møller’s hypothesis. The predictions are that (i) recordings from major cod spawning grounds should reveal sound with characteristics typical of cod whereas much less sound should be revealed outside the spawning season, (ii) calls from individual cod should show considerable variation, and (iii) calls from Arctic and Coastal cod should differ.

Material and methods

To study sound at a major spawning ground, recordings were carried out during the night at five stations off the Lofoten Islands at 6813.0’N 1438.7’E in Northern Norway, during the spawning season 8. and 9. April 1997, and half a year later on 4. September 1997 (Nordeide & Kjellsby 1999). The measuring hydrophone with a 32 dB gain built-in preamplifier had a total sensitivity of –152 dB re 1 V/mPa within the frequency range of 16 Hz to 2 kHz. In order to emphasize the transient character of the sound the digital recordings were analysed with Short-time Fourier Techniques (STFT), at the Norwegian Defence Research Establishment (Nordeide & Kjellsby 1999).

To compare the calls from the two groups, recordings were carried out in land-based tanks in 1998 to 2001 (Finstad, in prep.). Speciemens of the Arctic and Coastal cod were caught by trawl and transferred to tanks where recordings were carried out during the spawning period in 1998 - 2001. The smallest male used was 53 cm and the largest male was 94 cm long, whereas the smallest and largest females were 51 cm and 104 cm long, respectively. The average length of the five cod groups varied from 74.2 cm to 84.0 cm for males, and from 68.8 cm to 90.5 cm for females. Most recordings were from three 6 m diameter fibreglass tanks, but a 3 m diameter fibreglass tank was also used. Water level was 1.4 – 1.5 m in all tanks. Recording equipment was a 1 inch piezoceramic spherical hydrophone with a sensitivity of –198 dB ref 1V 1mPa, a Levell preamplifier type TA 601 with 60 dB gain, and a Sony TCD-D100 digital tape recorder. In 1998, the recordings were carried out with 12 speciemens (six Coastal males) in the experimental tank, whereas seven Arctic cod (5 males) where present in 1999. After the first years of experience we had identified two major problems: (i) we were not able to identify which cod produced the calls, and (ii) relatively few grunts had been recorded. In 2000 and 2001 we therefore chose to first record grunts with all cod in each group kept together, to increase the number of grunts. Thereafter, we split the groups of fish into a total of 10 smaller sub-groups to increase the minimum number of individual cod which could possibly produce the grunts. The groups in 2000 and 2001 consisted of 8 (3 males), 25 (19 males) and 22 (7 males) cod, respectively. The sub-groups consisted of from 4 (2 males) to 16 (11 males) cod. Towards the end of the spawning season the fish were killed. Examination of their otoliths by the Institute of Marine Research in Bergen, revealed that the recordings were carried out with groups and sub-groups consisting of (i) only male Coastal cod with or without the presence of Arctic females, (ii) only male Arctic cod with or without the presence of Coastal females, and (iii) a mixture of Coastal and Arctic males and females. These three alternative combinations are referred to as "Coastal-vocal", "Arctic-vocal" and "Mix-vocal" groups respectively, since only males produce sound during the spawning period (Brawn 1961c, Hawkins & Rasmussen 1978, see also Templeman & Hodder 1958, Engen & Folstad 1999). The number of individual cod (statistical "N") which could have produced the recorded grunts were minimum 3 and maximum 12 in the "Coastal-vocal" group, minimum 2 and maximum 7 in the "Arctic-vocal", and minimum 4 and maximum 22 in the "Mix-vocal" groups. Analysis by Avisoft-SASLabPro v. 3.74 provided estimates of harmonic spacing and duration of the grunts. Temporal structure of 78 recorded high quality grunts were analysed from oscillograms. Parameters included were number of downward peaks, time-intervals between peaks, and duration of the grunt. These parameters were analysed by Principal Component Analysis by the software "The Unscrambler" v. 7.5.

Results and Discussion

Field recordings provide support for the hypothesis that acoustic communication is important during cod spawning. Sound recordings at the major spawning ground off the Lofoten Islands revealed a hushed hubbub of sound, at approximately 40 – 500 Hz during the spawning period (Fig. 1a). Much less sound was revealed in September (Fig. 1b) when no cod spawn and migratory cod had emigrated to the Barents Sea. Nordeide & Kjellsby (1999) argues that this sound most likely is made by spawning cod since (1) The sound activity is highest in the frequency range where it has been suggested cod communicate (Chapman & Hawkins 1973; Hawkins & Rasmussen 1978). (2) The sound above 50 Hz had a transient character as expected for cod grunts. (3) More than ca. 50 million male cod spawned off the Lofoten Islands in April 1997, and recordings were made where the Institute of Marine Research in Bergen located the highest densities of spawning cod. (4) Cod totally dominated, by constituting more than 98 % by wet weight, the experimental and commercial catches in the area the seven days before, during and after the recordings (Nordeide & Kjellsby 1999).

Variation in calls between individual cod kept in tanks is as expected under the hypothesis that acoustic communication is important during female mate choice. Grunts from cod kept in tanks vary in harmonic spacing from 42 Hz to 79 Hz, and in duration from 0.11 s to 1.25 (Table 1). The two grunts with the lowest and highest harmonic spacing came from two different individuals, since they were recorded from two different sub-groups. The shortest and longest calls were also produced by two different individuals. This shows that different cod individuals may grunt at different frequencies and durations, but we cannot tell anything about each individual cod’s possibility to vary their calls.

The average harmonic spacing of grunts from the Coastal-vocal groups and Arctic-vocal groups were 53.4 Hz and 55.7 Hz, respectively, and the calls from the two cod groups lasted on average 0.33 s and 0.31 s, respectively (Table 1). The difference of 2.3 Hz in harmonic spacing and 0.02 s in duration between the two groups, is probably negligible. The difference cannot be tested statistically because the grunts are not independent events, since we were not able to tell which cod produced each call. Moreover, the calls from the Mix-vocal groups show no bimodal distribution in harmonic spacing or duration, as is expected if Coastal and Arctic cod call at two separate frequencies or durations. In the multivariate analysis of the temporal structure of the grunts, the first and second principal component explained 45 and 20%, respectively, of the total variation. However, the analysis did not cluster the grunts from Arctic-vocal and Coastal-vocal cod into two separate groups, as should be expected if the temporal structure of the two cod-groups differed (Finstad, in prep.). The hypothesis thus failed to pass the third test, since we have not been able to separate the calls from the two cod groups. However, analysis of temporal structure will continue with less rough analytical tools.

Bergstad, O.A., T. Jørgensen, O. Dragesund 1987. Life history and ecology of the gadoid resources of the Barents Sea. – Fisheries Research 5:119-161.

Brawn, V.M. 1961c. Sound production by the cod (Gadus callarias L.).- Behaviour 18:239-255.

Chapman, C.J. & A.D. Hawkins 1973. A field study of hearing in the cod, Gadus morhua L. – Journal of Comparative Physiology. 85: 147-167.

Engen, F. & I. Folstad 1999. Cod courtship song: a song at the expense of dance?- Canadian Journal of Zoology 77:542-550.

Finstad, J.L. In prep. Acoustic calls of Norwegian coastal cod and North-east Arctic cod (Gadus morhua). Cand. scient. thesis at the Departure of Fisheries and Marine Biology, University of Bergen. (Planned to be delivered in May 2002).

Hawkins, A.D. & K.J. Rasmussen 1978. The calls of Gadoid fish. – Journal of the Marine Biology Association of U.K. 58:891-911.

Hutchings, J.A., T.D. Bishop & C.R. McGregor-Shaw 1999. Spawning behaviour of Atlantic cod, Gadus morhua: evidence of mate competition and mate choice in broadcast spawner.- Canadian Journal of Fisheries and Aquatic Sciences 56:97-104.

Hylen, A. 1964. Coastal cod and skrei in the Lofoten area. - Fiskeridirektoratets Skrifter Serie Havundersøkelser, 13:27-42.

Møller, D. 1968. Genetic diversity in spawning cod along the Norwegian coast.- Hereditas 60:1-32.

Nordeide, J.T. 1998. Coastal cod and north-east Arctic cod – do they mingle at the spawning grounds in Lofoten?- Sarsia 83:373-379.

Nordeide, J.T. & I. Folstad 2000. Is cod lekking or a promiscuous group spawner?- Fish and Fisheries 1:90-93.

Nordeide, J.T. & E. Kjellsby 1999. Sound from spawning cod at their spawning grounds.- ICES Journal of Marine Science 56:326-332.

Nordeide, J.T. & I.H. Pettersen 1998. Haemoglobin frequencies and vertebrae numbers of cod (Gadus morhua L.) off northern Norway – test of a population structure hypotheses.- ICES Journal of Marine Science 55:134-140.

Rollefsen, G. 1934. The cod otolith as a guide to race, sexual development and mortality.- Rapports et Proces-Verbaux des Reunions du Conseil International Pour l’Exploration de la Mer 88:1-6.

Rollefsen, G. 1954. Observations on the cod and cod fisheries of Lofoten.- Rapports et Proces-Verbaux des Reunions du Conseil International Pour l’Exploration de la Mer, 136:40-47.

Templeman, W. & V.M. Hodder 1958. Variation with fish length, sex, stage of sexual maturity, and season in the appearance and volume of the drumming muscles of the swimbladder in the haddock, Melanogrammus aeglefinus (L.).- Journal of the Fisheries Research Board of Canada. 15:355-390.

Table 1
Mean, minimum and maximum values of frequency and duration of grunts from CC-vocal1, NAC-vocal2 and the MIX-vocal3 group. Number of grunts (N) of frequency measurements and duration measurements differ because some of the duration measurements were difficult to perform due to background noise. The table is from Finstad (in prep.).

Table of values of frequency and duration of grunts

1 : Grunts from groups and sub-groups consisting of CC males and CC and NAC females, based on otolith analyses
2 : Grunts from groups and sub-groups consisting of only NAC individuals, based on otolith analyses
3 : Grunts from groups and sub-groups consisting of both CC and NAC, both males and females, based on otolith analyses


graph of recordings from a major spawing ground
Figure 1. Recordings from a major spawning ground off the Lofoten Islands, (a) during the spawning period in April and (b) in September when no cod spawn. Reprinted from ICES Journal of Marine Science,Vol. 56, Nordeide, J.T. & E. Kjellsby, Sound from spawning cod at their spawning grounds, 326-332, 1999, by permission of the publisher Academic Press.


Return to Top | Workshop Proceedings: Short Papers