1 Introduction
During the past decade the international project GEOTRACES has undertaken an unprecedented effort to improve our knowledge of the distribution and biogeochemical cycles of trace elements and their isotopes (TEI) in the ocean (Anderson 2020). Trace metals are required in many metabolic functions (Sunda 2012) and as such, biogenic particles that are generated in the upper ocean are one of the main players regulating the internal cycling of so called “bioactive” trace elements (TE). TE are incorporated into particles through biological uptake and/or passive adsorption; they can then be remineralized, desorbed and/or ultimately exported. Particle sinking is one of the mechanisms of downward transport of chemical elements (Boyd et al. 2019). Large biogenic particles such as phytoplankton aggregates or fecal pellets are major vectors in the downward transport of particulate organic carbon (POC), but lithogenic particles can also play a role by ballasting aggregates and reducing remineralisation (Lemaitre et al. 2020). Due to their role in the control of atmospheric CO2 (Antia et al. 2001), carbon vertical export fluxes have been extensively studied, yet TE export fluxes have been considerably less investigated.
Vertical fluxes of particulate TE can be determined in different ways (McDonnell et al. 2015). Among them, moored sediment traps have been used since the 1970s to measure vertical fluxes of sinking material in the ocean. Important characteristics of vertical fluxes were revealed by this approach, but possible biases and limitations were also identified leading to the delivery of best practises (Buesseler et al. 2007). Taking these recommendations into account for deployment of the moorings and their design, moored sediment traps are powerful tools and are quite unique to capture long term variability (months to years) of sinking fluxes including TEs (Kremling and Streu 1993; Huang and Conte 2009; Kuss et al. 2010; Conte et al. 2019). In the northern Sargasso Sea, the oceanic flux program (OFP) provides the longest time series for elemental composition of export fluxes at three depths. Data collected between 2000 and 2015 were used to build mean seasonal cycles for 19 elements with a monthly temporal resolution (Conte et al. 2019). Based on the assumption that the elemental composition of the upper continental crust approximated lithogenic material composition in the traps, elemental fluxes were partitioned between different phases (organic matter, carbonates, lithogenic, authigenic). Two components have been identified as main drivers of the seasonal dynamics of the elemental fluxes. One of them was coupled to the seasonal cycle of primary production and surface export. The other one was related to internal processes associated with chemical scavenging and particle aggregation (Conte et al. 2019).
Based on a 13 year time series of elemental flux composition at 2000 m in the North East Atlantic, the mean annual cycles (monthly resolution) of 13 elements were reported (Pullwer and Waniek 2020). Overall, depending on the element considered, weak or no seasonality was detected. This was likely due to a small biological signal at the seasonal level, which was further damped by interannual variability of environmental conditions in surface waters. The depth of the traps also likely contributed to mask a clear seasonal signal at this site.
In the Southern Ocean, moored sediment traps were also widely used to investigate carbon export dynamics (Honjo et al. 2000). However, studies of the seasonal dynamics of TE export are rare. The seasonality of particulate export fluxes of 8 TE was studied in the polynya of Pridz Bay (Sun et al. 2016). Seasonal variations of Cu, Zn and Cd were mainly driven by ice coverage and biological production whereas fluxes of Al, Fe and Mn mainly derived from continental debris were controlled by ice melting and freezing processes.
Further investigations of the seasonal export dynamics of TE, with high temporal resolution are therefore required. We have addressed this challenge in a productive region of the Southern Ocean, the Kerguelen plateau where iron fertilisation leads to a marked seasonal pattern of carbon export (Rembauville et al. 2015a; Blain et al. 2020). In this context, we aimed to study and understand the various processes impacting the stoichiometry and the magnitude of these export fluxes, including the seasonal dynamics that can facilitate the partitioning of TE export between different carrier phases.