Interactions between behaviour and physical forcing in the control of horizontal transport of decapod crustacean larvae

We summarize what is known of the biophysical interactions that control vertical migration and dispersal of decapod larvae, asking the following main questions: How common is vertical migration in decapod crustacean larvae? What is the vertical extent of the migrations? What are the behavioural mechanisms that control vertical migrations? How does vertical migration interact with the physics of the ocean to control the dispersal of larvae? These questions are analysed by first giving a synopsis of the physical processes that are believed to significantly affect horizontal transport, and then by describing migration patterns according to taxon, to ecological category based on the habitat of adults and larvae, and to stage within the larval series. Some kind of vertical migration has been found in larval stages of virtually all species that have been investigated, irrespective of taxonomic or ecological category. Most vertical migration schedules have a cyclic nature that is related to a major environmental cyclic factor. Tidal (ebb or flood) migration and daily (nocturnal and twilight) migration are the two types of cyclic migration that have been identified. In general, all species show some type of daily migration, with nocturnal migration being the most common, whereas tidal migrations have only been identified in species that use estuaries during part of their life cycle. Moreover, there are several examples indicating that the phasing and extent of migration both change throughout ontogeny. Reported ranges of vertical displacement vary between a few metres in estuaries and several tens of metres (sometimes more than 100 m) in shelf and oceanic waters. Vertical movements are controlled by behavioural responses to the main factors of the marine environment. The most important factors in this respect are light, pressure and gravity, but salinity, temperature, turbulence, current and other factors, also influence behaviour. Many of these factors change cyclically, and the larvae respond with cyclic behaviours. The type of response may be endogenous and regulated by an internal clock, as in the case of some tidally synchronised migrations, but in most cases it is a direct response to a change in an environmental variable, as in diel migration. The reaction of the larvae to exogenous cues depends both on the rate of change of the variable and on the absolute amount of change. A series of dispersal types, involving different spatial and temporal scales, have been identified in decapod larvae: retention of the larval series within estuaries; export from estuarine habitats, dispersal over the shelf, and reinvasion of estuaries by the last stage; hatching in shelf waters and immigration to estuaries by late larvae or postlarvae; complete development on the shelf; and hatching in shelf waters, long-range dispersal in the ocean, and return to the shelf by late stages. In all of these cases, vertical migration behaviour and changes of behaviour during the course of larval development have been related to particular physical processes, resulting in conceptual mechanisms that explain dispersal and recruitment. Most decapod larvae are capable of crossing the vertical temperature differences normally found across thermoclines in natural systems. This ability may have significant consequences for horizontal transport within shelf waters, because amplitude and phase differences of the tidal currents across the thermocline may be reflected in different trajectories of the migrating larvae