Biogenic habitats play important roles in shallow-water ecosystems, but their roles in deeper waters are less well-studied. We quantitatively assessed 19 glass sponge reefs in the Salish Sea for live reef-building sponge cover and biodiversity, explored potential drivers behind variation observed among reefs, and quantified individual and collective roles the reefs play in filtration and carbon removal. The reefs support diverse and abundant communities of invertebrates and fish, with 115 unique taxonomic groups observed. Sponge cover varied widely between reefs: percent live reef-building sponge cover ranged from 0.2 to 17.5% and proportion of live reef habitat category ranged from 0.2 to 92%. These differences were predominantly driven by the seabed terrain characteristics such as seafloor rugosity, curvature, and depth; human pressure measures explored in this study - density of anthropogenic objects and fishing footprint over the past 17 years - did not mask the natural influence of seabed terrain. The difference in sponge cover between the reefs led to wide variation in ecosystem function with individual reefs processing between 465 and 47,300 L/m per day. Collectively, each day the 19 reefs filter 1.04 × 10 L of water which corresponds to 1% of the total water volume in Strait of Georgia and Howe Sound combined. The reefs remove up to 1 g of carbon per m per day, comparable to carbon sequestration rates reported for terrestrial old growth forests and to "blue carbon" sequestration rates by marine vegetation. Implications for sponge reef conservation and monitoring are discussed.

The largest known glass sponge reefs in Canada are within the Hecate Strait and Queen Charlotte Sound Glass Sponge Reefs Marine Protected Area (HSQCS-MPA) in British Columbia. However, human activities outside the core MPA boundaries, such as trawling, can create plumes of suspended sediments capable of travelling large distances. We studied the response of 3 glass sponge species to changes in suspended sediment concentrations (SSCs) at 170 m depth inside the HSQCS-MPA. Two species reduced excurrent flow rate in response to natural and experimentally induced increases in suspended sediment. Background suspended sediment levels were low and showed little variation (2.71 ± 0.09 mg l −1 , mean ± SD). Species varied in the threshold of SSCs that triggered arrests. Sediment concentrations of 2.8− 6.4 mg l −1 caused arrests in Rhabdocalyptus dawsoni, while Heterochone calyx did not arrest until concentrations reached 5−10 mg l −1. Very small, but prolonged increases of suspended sediments (<1 mg l −1 for R. dawsoni and 3.2 mg l −1 for H. calyx) caused arrests of several hours in R. dawsoni and prolonged periods of reduced flow in H. calyx. No arrests were observed in Farrea occa, even after repeated exposures up to 57 mg l −1. A sediment transport model showed that sediment concentrations can remain high enough to affect sponge behaviour as far as 2.39 km from the source of the plume. The results highlight the importance of understanding the biology of different species for establishment of adequate boundaries in MPAs.

No abstract

No abstract