Diel and Lunar Variations in Larval Fish Supply in Malindi Marine Park , Kenya

Larval fish supply to reefs influences the structure of adult populations and their response to exploitation. Despite this, few studies have examined patterns of larval fish supply to reefs in the Western Indian Ocean. The temporal variability in fish larval occurrence was thus studied in Malindi Marine Park, Kenya, to assess diel and lunar patterns of larval replenishment. Monthly and 24-hour sampling for fish larvae were undertaken between March 2005 and March 2007. Replicate samples were obtained by towing a 2 m long plankton net (500 μm mesh size, 0.2 m2 mouth area) for six minutes within the park. Larval abundance (larvae.100 m-3± SE) in the park was significantly higher during spring tides (951± 408) than neap tides (394 ± 260). Nocturnal larval abundance was 13-fold greater than daytime concentrations, regardless of tidal regime. The dominant fish families in the larval pool were Engraulidae, Labridae, Blenniidae and Gobiidae. Spectral time-series analysis revealed that larval fish supply occurred in a 30-day cyclical pattern associated with the new moon and was synchronised in the long-term with the northeast monsoon season. Corresponding author: JMM Email: babaallan@yahoo.com INTRODUCTION Elucidation of the patterns and processes associated with the supply of fish larvae to habitats is critical to understanding the replenishment and structure of their fish populations (D’Alessandro et al., 2007). Larval supply to habitats is affected by stochastic factors such as chemical, physical and biological oceanographic conditions during planktonic phases (Richards & Lindman, 1987; Kaunda–Arara et al., 2009), and by more deterministic factors such as spawning regimes and seasonal cues (Johannes, 1978; Valles et al., 2001). It is unlikely that biological or physical factors will exclusively control variability in the larval supply to reefs. However, larvae have been reported to frequently settle on a cyclical basis, correlated with short-term factors such as lunar cycles or tidal amplitude (Johannes, 1978; Taylor, 1984; Leis, 1993; D’Alessandro et al., 2007; Kaunda-Arara et al., 2009), and with large-scale factors such as inter-annual variability in ocean conditions (Doherty, 1987). In terms of the lunar cycle, greater larval settlement onto reef sites has been associated with new moon phases (Dufour & Galzin, 1993), perhaps as a strategy to avoid predation (Johannes, 1978). Environmental factors such as wind direction and speed can influence larval settlement at different temporal scales (Dufour & Galzin, 1993) and have been reported to affect larval distribution in nearshore habitats (Jenkins et al., 1998). Despite the effects of currents and wind on larval transport, some late-stage larvae of fish are known to actively control their position and dispersal (Leis, 1993). The composition, structure and timing of fish larval settlement from distant sites on reefs probably influence the population dynamics of the settled adult populations and their response to exploitation. Reefs with a high and diverse supply of fish larvae are likely to be less susceptible to and recover faster from the effects of overfishing (D’Alessandro et al., 2007). Therefore, it is important to understand both the temporal and the spatial scales of larval supply to these habitats. Moreover, larval transport between reef sites influences levels of connectivity and hence the genetic differentiation of reef populations (Sinclair, 1988; Botsford et al., 2009). Despite this, few studies have examined temporal and spatial variability in larval fish supply in the Western Indian Ocean (WIO) (but see Kaunda-Arara et al., 2009). Patterns in larval supply to reefs in the WIO at fine temporal scales are virtually unknown. The temporal variability in fish larval occurrence was thus studied in Malindi Marine Park, Kenya, to assess diel and lunar patterns of larval abundance. METHODS