Modeling the effects of temperature on the survival and growth of North Sea cod (Gadus morhua) through the first year of life

Abstract

Temperature and body size are widely agreed to be the primary factors influencing vital rates (e.g., growth, mortality) in marine fishes. We created a biophysical individual-based model which included the effects of body size and temperature on development, growth and mortality rates of eggs, larvae and juveniles of Atlantic cod (Gadus morhua L.) in the North Sea. Temperature-dependent mortality rates in our model were based on the consumption rate of predators of cod early-life stages. The model predicted 35%, 53% and 12% of the total mortality to occur during the egg, larval and juvenile stages, respectively. A comparison of modeled and observed body size suggested that the growth of survivors through their first year of life is high and close to the growth rates in ad libitum feeding laboratory experiments. Furthermore, our model indicates that experiencing warmer temperatures during early life only benefits young cod (or theoretically any organism) if a high ratio exists between the temperature coefficients for the rate of growth and the rate of mortality. During the egg stage of cod, any benefit of developing more rapidly at warmer temperatures is largely counteracted by temperature-dependent increases in predation pressure. In contrast, juvenile (age-0) cod experiences a higher cumulative mortality at warmer temperatures in the North Sea. Thus, our study adds a new aspect to the 'growth-survival' hypothesis: faster growth is not always profitable for early-life stages particularly if it is caused by warmer temperatures.

Oceanographic drivers of sablefish recruitment in the California Current

“…Rapid growth may help individuals avoid predation through a number of mechanisms (e.g., “bigger‐is‐better”, stage‐duration) that cannot be untangled in this study. However, Sogard () did not find evidence for size‐selective mortality, possibly because the higher temperatures also lead to increased predator activity and consumption (e.g., Akimova, Hufnagl, Kreus, & Peck, ). One might also expect the offshore transport associated with cold, upwelled waters to advect sablefish larvae away from suitable settlement habitat.…”

Predation on Atlantic herring (Clupea harengus) eggs by the resident predator community in coastal transitional waters

“…The predation on herring eggs is probably lower during earlier periods of the spawning season, where spawn concentrations are usually higher, the abundance of predators is relatively low (Kotterba et al ) and predator appetites are lower due to relatively cold water temperatures (e.g., see Akimova et al ). Our experiment was run during the second half of the herring spawning season when predators were likely more active.…”

Influence of larval transport and temperature on recruitment dynamics of North Sea cod (Gadus morhua) across spatial scales of observation

“…This might be due to the limited outflow and high retention in the South population. The observed effect of SST on recruitment could therefore be due to other mechanisms: for example food availability (Nicolas et al., 2014), physiological constraints (Butzin & Pörtner, 2016;Nunez‐Riboni, Taylor, Kempf, Pu, & Mathis, 2019) and predation by warm‐water predators (Akimova, Hufnagl, et al, 2016). For the Viking population, our results suggest that SST could influence recruitment through drift both positively and negatively.…”

“…During the simulation, eggs were subject to a fixed daily mortality rate ( m ) of 0.2, which is within the range of mean values estimated in studies of cod eggs (0.1–0.32, Rijnsdorp & Jaworski, 1990; see Table 2 in Langangen, Stige, Yaragina, Vikebø, et al, 2014). For larvae, we set the mortality rate to decrease with weight (Figure 2e) as parameterised for North Sea cod larvae in Akimova, Hufnagl, Kreus, and Peck (2016), based on the size‐spectrum theory (Peterson & Wroblewski, 1984): The survival probability of each individual was updated throughout the simulation according to the mortality rate (i.e. individuals were not removed from the simulation), following a super‐individual approach (Scheffer, Baveco, DeAngelis, Rose, & van Nes, 1995).…”