Elevated mortality of fish larvae on coral reefs drives the evolution of larval movement patterns

Abstract

Coral reef fishes typically undergo a pelagic larval phase prior to recruitment to reef habitat. This is potentially risky, but likely to be important in connecting populations and avoiding local crowding. Predation pressures on larvae on and off the reef are likely to differ both in origin and intensity. In this study, we used individual-based models to explore in isolation the effect of elevated levels of larval mortality on the reef in 22 different landscapes. We allowed the movement of highly simplified larvae to evolve through selection and mutation events and tracked emerging strategies over 180 generations. The pressure of increased larval mortality on the reef affected overall population sizes and selected for higher levels of larval movement. However, the evolution of movement was constrained, and self-recruitment back to the reef of origin was prevalent, with levels rarely dropping below 50%. The evolved strategies were highly landscape specific, suggesting that movement in contiguous reefs is more readily evolved than at isolated reefs. Future development of this simulation approach will provide a valuable research tool for exploring important evolutionary, ecological, and management-based questions.KEY WORDS: Evolved strategy · Larvae · Coral reef fishes · Active dispersal · Self-recruitment Resale or republication not permitted without written consent of the publisher

When is dispersal for dispersal? Unifying marine and terrestrial perspectives

“…Many factors influence this trade-off, such as temperature, developmental time and survival rate, fertilization success, the ability to feed facultatively, generation time, and post-metamorphic effects of offspring size (reviewed in Marshall & Keough, 2008). Larval feeding and growth in the plankton, as well as reduced benthic mortality during early developmental stages, represent separate benefits from those directly related to dispersal, but could affect selection on fecundity, initial size, and size at settlement, which in turn influence pelagic larval duration and patterns of dispersal (Baskett et al, 2007;Dytham & Simpson, 2007Pringle et al, 2014. Realized dispersal also depends on the timing of offspring release into the water.…”

“…For example, general theory (Dytham & Simpson, 2007) and empirical evidence from terrestrial systems (Cody & Overton, 1996) demonstrate selection against dispersal on islands, but many island endemic reef fishes have planktonic larvae with dispersal potentials similar to those of more widespread species (Robertson, 2001). In terms of reducing dispersal, long-term constraints on the evolution and loss of capacity for larval feeding may limit selection on larval duration as a means of adjusting dispersal (Pechenik, 1999;Bonhomme & Planes, 2000).…”

“…Weighing costs of dispersal, such as planktonic mortality or transport to unsuitable habitat, against the nutrition and survivorship benefits of a pelagic versus benthic larval stage, an offspring size-number trade-off can favour the evolution of a pelagic stage as the optimal strategy (Vance, 1973;Baskett et al, 2007;Pringle et al, 2014). Indeed, it may be adaptive for larvae to move away from parental habitat to develop, but maladaptive for larvae to disperse too widely and settle in distant locations (Table 1; Johannes, 1978;Strathmann et al, 2002;Baskett et al, 2007;Dytham & Simpson, 2007).…”

Evolutionary optimality in stochastic search problems

“…One key simplification is the selection of a new search direction for each movement step that is independent of any previous movement direction, and this leads to the models being classed as zero-information searches. Although there is experimental evidence to support larval foragers exhibiting rudimentary cognitive ability (Dytham & Simpson 2007), the zero-information assumption is commonly applied in foraging simulations (Zollner & Lima 1999;Viswanathan et al 2000;Benhamou 2007) and we continue with that assumption here.…”

Preventing overexploitation of migratory fish stocks: the efficacy of marine protected areas in a stochastic environment

“…However, long-term changes in habitats (e.g. increased natural predation within an MPA) may favour migration away from those areas (Dytham and Simpson, 2007). Such detailed evolutionary considerations are beyond the scope of this model.…”