Nitrogen cycling in deeply oxygenated sediments: Results in Lake Superior and implications for marine sediments

To understand the nitrogen (N) cycle in sediments with deep oxygen penetration, we measured pore-water profiles to calculate N fluxes and rates at 13 locations in Lake Superior in water depths ranging from 26 to 318 m. Sediments with high oxygen demand, such as in nearshore or high-sedimentation areas, contribute disproportionally to benthic N removal, despite covering only a small portion of the lake floor. These sediments are nitrate sinks (average 0.16 mmol m22 d21) and have denitrification rates (average 0.76 mmol m22 d21) that are comparable to those in coastal marine sediments. The deeply oxygenated (4 to . 12 cm) offshore sediments are nitrate sources (average 0.26 mmol m22 d21) and generate N2 at lower rates (average 0.10 mmol m22 d21). Ammonium is nitrified with high efficiency (90%), and nitrification supports . 50% of denitrification nearshore and , 100% offshore. Oxygen consumption by nitrification accounts for 12% of the total sediment oxygen uptake. About 2% of nitrate reduction is coupled to oxidation of iron, a rarely detected pathway. Our Lake Superior N budget indicates significant contributions from sediment–water exchanges and N2 production and is closer to balance than previous budgets. Our results reveal that sediment N cycling in large freshwater lakes is similar to that in marine systems. They further suggest that denitrification rates in slowly accumulating, welloxygenated sediments cannot be described by the same relationship with total oxygen uptake as in highsedimentation areas; hence, global models should treat abyssal ocean sediments differently than coastal and shelf sediments. Transformations of nitrogen in aquatic sediments are an important part of the global nitrogen (N) cycle. Sediments actively exchange the reactive nitrogen—nitrate and ammonium—with the water column and contribute to its recycling and removal, accounting for 50–70% of the denitrification in the global ocean (Codispoti et al. 2001). Denitrification (reductive conversion of nitrate to dinitrogen) and nitrification (oxidation of ammonium to nitrate) are the most commonly considered reactions in the sediment nitrogen cycle, though anammox (anaerobic oxidation of ammonium to N2 with NO { 2 as the electron donor) is increasingly recognized as a significant pathway of nitrogen removal in marine sediments (Dalsgaard et al. 2005). As the reaction rates (Laursen and Seitzinger 2002) and sediment–water exchange fluxes of nitrogen in marine sediments are highly variable (Devol and Christensen 1993), for the purposes of the phenomenological descriptions and global N-cycle modeling sediments are often categorized based on water depth and environment, as follows: estuaries, bays, shelf and coastal oceans, deeper continental margins, and deep oceans (Middelburg et al. 1996; Fennel et al. 2009; Glud et al. 2009). To link the N transformation rates to commonly measured quantities, the rates of nitrification and denitrification have been correlated to sediment oxygen consumption, sedimentation rates, and water depth (Middelburg et al. 1996; Seitzinger et al. 2006). Whereas in marine environments such relationships have been described relatively well, nitrogen cycling in large freshwater lakes has received less attention. Lake Superior, the world’s largest lake by surface area, provides an opportunity to investigate the nitrogen cycle in a freshwater end member system that in many respects is similar to marine systems. Characterized by relatively low organic carbon (C) content (3–5 wt%; Li et al. 2012; Kistner 2013) and slow accumulation rates, the offshore sediments of Lake Superior exhibit an exceptionally deep penetration of oxygen (2 to . 16 cm in water depths between 120 and 300 m; Li et al. 2012), typical of oceanic hemipelagic sediments in 2000–3000 m water depth (Glud 2008). The organic carbon mineralization rates and carbon burial efficiencies in these sediments are similar to those in the deep ocean (Li et al. 2012). The nitrogen dynamics in Lake Superior have been enigmatic (Sterner et al. 2007), as over the last century the lake has experienced an unusual increase in water column nitrate concentrations, leading to an extreme N : P ratio of 10,000 (Guildford and Hecky 2000). The nitrate accumulation has been suggested to result at least partly from the ammonium oxidation in the lake (Finlay et al. 2007), and tentative links to the sediment N cycling have been suggested (Finlay et al. 2013; Small et al. 2013). The sediment’s role in the nitrogen cycling in Lake Superior is poorly quantified (Sterner et al. 2007; Li 2011; Small et al. 2013), with scarce geographical coverage and little information available on the rates of critical geochemical pathways, such as nitrification and denitrification. The sediments in Lake Superior exhibit strong temporal and spatial variability, with oxygen penetration varying seasonally by as much as 2 cm (Li et al. 2012), especially at locations with deep oxygen penetration, and with lateral heterogeneity on scales from tens to hundreds of meters (Li et al. 2012). This variability complicates * Corresponding author: [email protected] Limnol. Oceanogr., 59(2), 2014, 465–481 E 2014, by the Association for the Sciences of Limnology and Oceanography, Inc. doi:10.4319/lo.2014.59.2.0465