Multivariate benthic ecosystem functioning in the Arctic – benthic fluxes explained by environmental parameters in the southeastern Beaufort Sea

Abstract

Abstract. The effects of climate change on Arctic marine ecosystems and their biogeochemical cycles are difficult to predict given the complex physical, biological and chemical interactions among the ecosystem components. To predict the impact of future changes on benthic biogeochemical fluxes in the Arctic, it is important to understand the influence of short-term (seasonal to annual), long-term (annual to decadal) and other environmental variability on their spatial distribution. In summer 2009, we measured fluxes of dissolved oxygen, nitrate, nitrite, ammonia, soluble reactive phosphate and silicic acid at the sediment-water interface at eight sites in the southeastern Beaufort Sea at water depths from 45 to 580 m to address the following question and hypotheses using a statistical approach: (1) What is the spatial variation of benthic boundary fluxes (sink and source)? (2) The classical proxy of benthic activity, oxygen flux, does not determine overall spatial variation in fluxes. (3) A different combination of environmental conditions that vary either on a long-term (decadal) or short-term (seasonal to annual) scale determine each single flux. And (4) A combination of environmental conditions varying on the short and long-term scale drive the overall spatial variation in benthic boundary fluxes. The spatial pattern of the measured benthic boundary fluxes was heterogeneous. Multivariate analysis of flux data showed that no single or reduced combination of fluxes could explain the majority of spatial variation. We tested the influence of eight environmental parameters: sinking flux of particulate organic carbon above the bottom, sediment surface Chl a (both short-term), porosity, surface manganese and iron concentration, bottom water oxygen concentrations (all long-term), phaeopigments (intermediate-term influence) and Δ13Corg (terrestrial influence) on benthic fluxes. Short-term environmental parameters were most important for explaining oxygen, ammonium and nitrate fluxes. Long-term parameters together with Δ13Corg signature explained most of the spatial variation in phosphate, nitrate and nitrite fluxes. Sediment pigments and Δ13Corg levels in surficial sediments were most important to explain fluxes of silicic acid. The overall spatial distribution of fluxes could be best explained (57%) by the combination of sediment Chla, phaeopigments, Δ13Corg, surficial manganese and bottom water oxygen concentration. We conclude that it is necessary to consider long-term environmental variability in the prediction of the impact of ongoing short-term environmental changes on the flux of oxygen and nutrients in Arctic sediments. Our results contribute to improve ecological models predicting the impact if climate change on the functioning of marine ecosystems.