Post-eruption succession of macrofaunal communities at diffuse flow hydrothermal vents on Axial Volcano, Juan de Fuca Ridge, Northeast Pacific

Abstract

Abstract Hydrothermal vents harbor dense aggregations of invertebrate fauna supported by chemosynthesis. Severe tectonic events and volcanic eruptions frequently destroy vent communities and initiate primary succession at new vents on ridge-crest submarine lava flows. An eruption on Axial Volcano (∼1500 m depth), a seamount on the Juan de Fuca Ridge (JdFR) in the northeast Pacific Ocean, occurred in January 1998, which created new substratum and vents. This study examines the development of the macrofaunal vent assemblages associated with tubeworms (Ridgeia piscesae) at eight diffuse flow vents over the following 3 years. Biological collections by suction of lava surfaces also characterized “pre-tubeworm assemblages”. Coupled fluid sampling showed an overall decrease in temperature, sulphide, and sulphide-to-heat ratios over 3 years as well as large spatial variability across the new vents. We examined collections of pre-eruption diffuse flow vent assemblages at Axial Volcano to compare the stages of new community development to “mature” vents. Mature vent assemblages are characterized by two major community types dominated by limpets (Lepetodrilus fucensis) and alvinellid polychaetes (Paralvinella pandorae and/or P. palmiformis). The following post-eruption succession patterns emerged. First, R. piscesae tubeworms took up to 3 years to establish aggregations at the new vents, and the majority of pre-tubeworm assemblages were dominated by grazing polychaetes. Second, species colonized quickly and by 30 months after habitat creation >60% of Axial's species pool had arrived at the new vents; abundance at mature vents predicted colonization success with some notable exceptions. Third, shifts in species dominance occurred rapidly and by Year 3 new vent assemblages resembled mature, pre-eruption communities. In general, tubeworm assemblages were dominated by alvinellid polychaetes (P. pandorae and P. palmiformis) in the first 2 years post-eruption, with limpets (L. fucensis) becoming more numerous in Year 3. Fourth, successional trajectories corresponded to four habitat variables: the timing of R. piscease recruitment, vent age, maximum vent fluid sulphide-to-heat ratios, and maximum vent fluid temperatures. The two mature community types (dominance by limpets or alvinellid polychaetes) seem to relate in part to vent differences in fluid properties, with limpets more prevalent than alvinellids at diffuse vents with lower maximum sulphide-to-heat ratios and temperatures. We present a general model of post-eruption succession for Juan de Fuca Ridge diffuse flow vents. The model incorporates potential abiotic and biotic drivers of community development and identifies six successional stages: a pre-tubeworm assemblage, four tubeworm assemblage states determined by the relative dominance of alvinellid polychaetes and the limpet, and a final senescent stage when venting wanes. Future eruptions will allow for this model to be tested through direct observation and experimentation.