Interaction of offshore and inshore processes controlling settlement of brachyuran megalopae in Saco mangrove creek, Inhaca Island (South Mozambique)

It is usually considered that reinvasion of the estuarine habitat by crab l arvae that develop in the sea involves 2 different steps, which are control led by different factors: transport of larvae from the shelf towards the co ast and upstream movement until appropriate settlement habitats are found. In order to investigate reinvasion of the Saco mangrove creek (Inhaca Islan d, Mozambique) by crab megalopae, the timing and spatial distribution of se ttlement were related to major environmental forcing factors. Several sets of experiments were performed, with different specific objectives, using me galopal collectors. The total supply of brachyuran megalopae to the mangrov e creek over a 2 mo period responded to 2 forcing factors, the first relate d to cycles of tidal amplitude and the second to offshore wind stress. The response of settlement on the collectors to the tidal amplitude cycle was r eflected by clear maxima recorded on every spring tide. This contrasts with the results of previous studies analysing single species recruitment proce sses, which found a much more irregular pattern. This result was interprete d as a consequence of 2 factors: (1) the use of a cumulative number of tota l Brachyura, which increases the probability of detecting supply pulses, an d (2) a larger tidal range at Inhaca than at the locations where most previ ous studies took place. The results of this experiment suggest interaction of the stochastic effect of offshore wind stress, which maximises transport to the coastal area, with the deterministic effect of the tide and spring- neap cycles, which operate at the nearshore level to carry megalopae to spe cific mangrove areas. Settlement was also studied by deploying collectors i n 3 different strata of the mangroves, and the results showed a differentia l settlement distribution in favour of lower areas in the vicinity of the m angroves. This suggests subsequent juvenile migration to suitable areas fol lowing first moult. Different settlement stimuli were also tested in the fi eld by enclosing in the collectors a mixture of mangrove crabs and mangrove vegetal material. The results were inconclusive, as no significant differe nces were found; this was attributed to the effects of interference of stim uli between adjacent collectors and of proximity to mangroves.