Foraminifera associated with phytodetritus deposits at a bathyal site in the northern Rockall Trough (NE Atlantic): seasonal contrasts and a comparison of stained and dead assemblages

Assemblages of ‘live’ (rose Bengal-stained) benthic foraminifera were analysed in multicorer samples (0–1-cm sediment layer) collected during May and July 1998 at a bathyal (1920 m water depth) site in the northern Rockall Trough, NE Atlantic. A phytodetrital deposit of variable thickness, presumably derived from spring bloom material, was present only in July 1998. Two cores, one from May and one from July, were wet-sorted for all stained benthic foraminifera (the ‘entire live assemblage’). One additional core from each sampling period was analysed for stained calcareous and selected agglutinated species. The 63–125-, 125–150- and >150-μm fractions were sorted separately in order to facilitate comparison with previous studies. Population densities were very high (2379 individuals 10 cm−2) in the July core where >38% of stained individuals occurred in the well developed phytodetrital layer. They were much lower (827 individuals 10 cm−2) in the May core which was devoid of phytodetritus. These populations contained a wide range of higher taxa, including soft-shelled monothalamous forms. A total of 263 morphospecies were recognised in the two samples: 158 in May and 176 in July. Some species showed sharp increases in population density between May and July. These changes were particularly evident in the case of rotaliids, but less pronounced among agglutinated and allogromiid taxa. The dominant species in the July samples showed the strongest fluctuations; Eponides pusillus and Nonionella iridea had mean densities of 6.6 and 40 individuals 10 cm−2, respectively, in the two May samples compared to 161 and 326 individuals 10 cm−2 in July. These two species exhibited different ecological characteristics; E. pusillus (test size usually 60–120 μm) almost always occurred within phytodetrital aggregates, whereas N. iridea (test size usually 80–180 μm) was often enveloped by an agglutinated cyst or embedded within a lump of sediment. In E. pusillus, mean test size was significantly smaller in July 1998 than in May 1998, suggesting that reproduction and growth were occurring in the presence of phytodetritus. N. iridea exhibited the opposite pattern (significantly smaller in May), perhaps because reproduction was less closely linked to the presence of phytodetritus and therefore proceeded at a slower rate. Dead foraminiferal assemblages were dominated by calcareous taxa and were less diverse than the stained assemblages. The two main phytodetrital species (N. iridea and E. pusillus) were important elements in the dead assemblages (ranked first and third, respectively). Some other species (notably Cassidulina obtusa), however, were relatively much more abundant in the dead assemblage than in the stained assemblage (i.e. they have low live/dead ratios). Possibly, their main period of test production occurred after July, perhaps as part of a successional sequence following phytodetritus deposition in spring/early summer. Overall, our results add to the growing body of evidence that the abundance and distribution of some small epifaunal and shallow infaunal foraminiferal species is closely linked to the deposition of labile organic matter to the deep-seafloor and that this ‘phytodetrital signal’ can be preserved in the fossil record.