Skipped spawning is common for the Northeast Arctic cod in a life-history energy allocation model. ICES CM 2004/K:28

That iteroparous fish reproduce every year after they have become sexually mature is a common assumption in fisheries science. Experimental and field studies suggest, however, that mature fish commonly skip spawning, often but not only in relation with failing food resources. This study presents a life-history model based on optimal energy allocation for the Northeast Arctic cod. Ingested energy can either be allocated to growth or to storage for the next spawning period. This trade-off embodies lifelong patterns of growth, maturation, and reproduction. The allocation decision was optimised based on five individual states: age, length, size of energy stores, month of the year, and the current food availability. The model showed that skipped spawning was partly related to periods of low food intake (there was not sufficient energy to spawn), but that it also played a central role in life-history: spawning should be omitted because of the benefits that come from growth to a larger size. Skipped spawning was most common in the years just following maturation. The influence of several parameters was tested, and skipped spawning was more common when: 1) juvenile mortality was high; 2) mortality at the spawning grounds was low; 3) the energetic cost of migration was high; and 4) food intake was low. Introduction Why should long-lived iteroparous individuals skip reproduction after they have reached sexual maturity? These individuals have already undergone the physiological changes associated with maturation, and are thus, in principle, prepared for repeated and successful reproduction. To understand such patterns of reproduction it is essential to appreciate how the life-long trade-off between reproduction, growth, and survival affects an organism’s life history. From the organism’s point of view, these trade-offs require strategies for differentially allocating available resources to changes in size, fecundity, and energy stored for future actions. Reproduction is more often skipped when individual condition is poor. In fish, this has been documented for several species (Bagenal, 1957; Rijnsdorp, 1990; Kjesbu et al., 1991; Rideout et al., 2000). Instead of exacerbating already low energy reserves, the time and energy required for reproduction should be channelled to increase future successes. Additional factors may modify the optimal life-history strategy. Since skipped spawning results from the trade-off between current and future reproduction, we would expect that factors such as mortality, food intake, and migration costs would trigger evolution of life history strategies, and thus influence the frequency and pattern of skipped spawning. This study explores reproductive strategies in Atlantic cod (Gadus morhua) using a flexible life-history model. Cod is a repeat spawner, and the Northeast Arctic stock, which is the object of this study, performs annual migrations from the feeding grounds in the Barents Sea to spawning grounds off Lofoten. The annual migrations entail a direct cost in terms of energy and time. In the present model, we focus on this allocation rule, which is dependent on age, time of year, body length, state of the energy stores, and food availability. Each point on the resulting fivedimensional surface that describes the allocation rule is independent, so that the allocation rule is not artificially constrained but allows any shape to emerge evolutionarily. The optimal allocation rule, which is also the allocation rule supposedly favoured by natural selection, can then be found for various environmental scenarios. On this basis, patterns of growth and reproduction can be predicted and analyzed. Using the model thus outlined, this paper first focuses on the mechanisms of skipped spawning, then turns to factors that affect lifehistory strategies in an evolutionary perspective. Model description Our study is based on a flexible life-history model describing a migratory fish (a detailed description of the model will be found in Jørgensen and Fiksen, In prep.). The model is parameterized for the Northeast Arctic cod stock and fits well with observed growth in natural stocks experiencing variable environmental conditions. A key assumption of the model is that energy, ingested in a stochastic feeding environment, can be allocated either to growth or to storage (Fig. 1). Growth is irreversible and increases somatic structures together with a minimum amount of muscle mass. Energy stores (lipids in the liver and increased white muscle mass) grow reversibly between a minimum and maximum condition factor, and the stored energy can be utilized for metabolic or reproductive purposes. Eggs are spawned in one batch in March, and reproduction requires migrations to the spawning grounds. Migration is costly in terms of energy (increased metabolic rate during the migration), time (cod eat little during migration and spawning and thus forego opportunities for growth), and mortality (natural mortality increases during migration and spawning). In addition to natural mortality, fishing mortality can occur on the feeding (ZF) and spawning (ZS) grounds due to human fisheries. Optimal life-history strategies were calculated using dynamic optimization (Houston and McNamara, 1999; Clark and Mangel, 2000), using four state variables: age (in months, thus including seasonal variations), stored energy (measured on a relative scale between 0 and 1), body length (cm), and food availability. The optimal allocation strategies were then simulated and population dynamics for 1.000 years analyzed to derive consequences for individual and population patterns of growth, maturation, and reproduction. Three scenarios were used in our investigation of skipped spawning: 1) a moderate spawner fishery and a light fishery at the feeding grounds (ZS=0.1, ZF=0.05), resembling the fishing regime at the beginning of the 20 century and resulting in a life history with late maturation; 2) a high spawner fishery and an intense fishery on the feeding grounds (ZS=0.3, ZF=0.5), resembling contemporary fishing and resulting in early maturation; and Food intake Metabolism Migration Age and month Body length Stored energy Current food