Spatial and temporal variations of recruitment in the tube worm Riftia pachyptila on the East Pacific Rise (9°50'N and 13°N)

Abstract

The giant vestimentiferan tubeworm Riftia pachyptila is one of the most abundant and ecologically important species at deep-sea hydrothermal vents along the East Pacific Rise (EPR). Clumps of R. pachyptila could indeed be viewed as one of the most frequent habitats for the fauna typifying the eastern Pacific hydrothermal vent sites and thus deserve specific attention, as their dynamics are likely to control the spatial and temporal evolution of numerous vent species. As a con- sequence, population structure and recruitment of R. pachyptila were studied from samples taken at different vent localities during 2 EPR surveys in 1996 and 1997. In 1996, 3 samples were collected within and between the hydrothermal vent fields 13° N and 9° 50' N to assess the spatial coherence of recruitment. In 1997, a huge number of additional specimens were collected at 9° 50' N from a single vent site at 3 successive dates over 22 d to determine how growth rate may affect the temporal vari- ability of population structure at a small spatial scale. Size-frequency analyses were performed using the vestimentum width as a biometric index. Inter-site comparisons revealed that size-frequency dis- tributions did not differ significantly within a vent field suggesting the synchronisation of recruitment locally (i.e. a few km). By contrast, population structure strongly differed between the 2 vent fields in response to local variations in time of larval supply and/or reproductive activity. The size-frequency histograms of the 3 samples collected in 1997 displayed a regular polymodal structure reflecting a discontinuous recruitment in time. The presence of a large number of small individuals attached to the tube of conspecific larger ones suggested that biological cues associated with Riftia may promote habitat selection by settlers. Differences between the modal vestimentum width of each population 'cohorts' from the 3 temporal samples were significant and could result from either a growth rate het- erogeneity in response to local environmental variations or to population ageing. According to this 2nd assumption, growth rate was estimated from the evolution of the 5 'cohorts' found over 22 d. Our data suggested that R. pachyptila exhibits: (1) a high yearly tube growth rate (estimated to about 160 cm yr -1 ); (2) a high individual mortality rate; and (3) a high population turnover, all of which are the characteristics of an opportunistic species.