Pervasive sources of isotopically light zinc in the North Atlantic Ocean

Highlights

• Hydrothermal and sedimentary sources of light ${\delta }^{66}$Zn.

• ${\delta }^{66}$Zn and abundance ratios to the major nutrients deconvolve key processes.

• Light ${\delta }^{66}$Zn in the surface North Atlantic due to Zn addition, not scavenging.

Abstract

In this study, we report seawater dissolved zinc (Zn) concentration and isotope composition (${\delta }^{66}$Zn) from the GEOTRACES GA01 (GEOVIDE) section in the North Atlantic. Across the transect, three subsets of samples stand out due to their isotopically light signature: those close to the Reykjanes Ridge, those close to the sediments, and those, pervasively, in the upper ocean. Similar to observations at other locations, the hydrothermal vent of the Reykjanes Ridge is responsible for the isotopically light Zn composition of the surrounding waters, with an estimated source ${\delta }^{66}$Zn of -0.42 ‰. This isotopically light Zn is then transported over a distance greater than 1000 km from the vent. Sedimentary inputs are also evident all across the trans-Atlantic section, highlighting a much more pervasive process than previously thought. These inputs of isotopically light Zn, ranging from -0.51 to +0.01 ‰, may be caused by diffusion out of Zn-rich pore waters, or by dissolution of sedimentary particles.

The upper North Atlantic is dominated by low ${\delta }^{66}$Zn, a feature that has been observed in all Zn isotope datasets north of the Southern Ocean. Using macronutrient to Zn ratios to better understand modifications of preformed signatures exported from the Southern Ocean, we suggest that low upper-ocean ${\delta }^{66}$Zn results from addition of isotopically light Zn to the upper ocean, and not necessarily from removal of heavy Zn through scavenging. Though the precise source of this isotopically light upper-ocean Zn is not fully resolved, it seems possible that it is anthropogenic in origin. This view of the controls on upper-ocean Zn is fundamentally different from those put forward previously.

Introduction

Zinc (Zn) is an essential micronutrient for marine primary producers (Morel and Price, 2003). It is required for key metalloenzymes such as carbonic anhydrase, which is involved in carbon fixation, or alkaline phosphatase, which gives phytoplankton access to organic forms of phosphorus when phosphate concentrations are low (Sunda, 1989). As a result, the marine cycles of zinc and carbon are intrinsically linked.

Analytical advances over the last decade have enabled study of the stable isotope composition of Zn (${\delta }^{66}$Zn = variations in ⁶⁶Zn/⁶⁴Zn expressed in parts per thousand deviation from the JMC Lyon standard), to investigate the processes controlling the marine Zn distribution (Bermin et al., 2006; Conway et al., 2013; Takano et al., 2013). In addition, the recent international programme GEOTRACES has provided a large quantity of high-quality data, from full-depth profiles and sections, allowing new insights into the large-scale distribution of trace elements, including Zn (Conway and John, 2014, 2015; Zhao et al., 2014; Vance et al., 2016; John et al., 2018; Weber et al., 2018; Wang et al., 2019). However, gaps remain in our understanding of the modern Zn cycle. Firstly, the ${\delta }^{66}$Zn of seawater (averaging +0.46 ‰) is higher than the known inputs and lower than most known outputs, pointing to a missing budget term if the oceanic Zn cycle is in steady state (Little et al., 2014; Moynier et al., 2017). Secondly, north of the Southern Ocean, a shift toward light Zn isotope signatures in the dissolved pool is observed within the upper ocean (<1000 m). This is surprising, given that isotopic fractionation between phytoplankton cells and the dissolved pool is thought to be close to zero, or slightly in favour of light isotope uptake, which should leave the residual dissolved pool slightly heavy (John et al., 2007; Peel et al., 2009; Samanta et al., 2017; Köbberich and Vance, 2019; Wang et al., 2019). Laboratory experiments have suggested that Zn released from degrading phytoplankton cells can be rapidly scavenged back onto organic matter, and that this adsorbed Zn is isotopically heavier than the dissolved pool (John and Conway, 2014). Scavenging of isotopically heavy Zn onto sinking biogenic particles has thus been suggested to explain the low ${\delta }^{66}$Zn values in the upper ocean (Conway and John, 2014, 2015; John et al., 2018; Weber et al., 2018), although it should be noted that a dominant proportion of marine dissolved Zn is complexed to natural organic ligands (e.g. Ellwood and Van den Berg, 2000) and thus presumably not available for adsorption to particles.

Here we examine Zn isotopes and concentrations along a GEOTRACES section that crosses the North Atlantic from the Iberian Peninsula to Newfoundland (Fig. 1). The North Atlantic is a promising area to study biological, physical and geochemical processes affecting micronutrient distributions, as it is characterised by a strong spring bloom (Longhurst, 2010), the formation of globally-important deep water masses (e.g. Daniault et al., 2016), and a variety of trace metal sources (Ohnemus and Lam, 2014). In this study, we focus our discussion on the processes responsible for the light isotope composition of Zn observed at the Reykjanes Ridge, at the sediment-water interface, and in the upper 500 m of the ocean. In doing so, we expand our analysis to data from the entire North Atlantic. We combine macronutrient/Zn ratios with Zn stable isotope data for the dissolved pool in order to better identify the processes that modify preformed Southern Ocean signatures in the low-latitude oceans. In contrast to previous studies that invoke scavenging removal of heavy Zn isotopes for the origin of light upper-ocean Zn (e.g. John and Conway, 2014), we conclude that the Zn isotope signature of the upper ocean is dominated by the addition of isotopically light Zn to upper-ocean water masses, whose preformed Zn concentrations are extremely low (Vance et al., 2017; de Souza et al., 2018; Middag et al., 2019), and which are thus very sensitive to the addition of small amounts of Zn.