Using stable isotopes as tracers of water masses and nutrient cycling processes in the Gulf of Maine

Highlights

• Differences in d¹⁸O_water in different Gulf of Maine basins suggest differences in vertical mixing and freshwater input.

• d¹⁵N_NO3- and d¹⁸O_NO3- values of Gulf of Maine water samples suggest significant influence of phytoplankton assimilation.

• The Δ(15, 18) variable suggests nitrification is occurring in the upper water column throughout much of the Gulf of Maine.

Abstract

The dramatic marine environmental change seen today can be difficult to fully document and interpret without adequate, spatially and temporally comprehensive, baseline datasets of hydrographic properties. Here we present isotope data measured in water samples collected during a nine-day research cruise in October 2016 throughout the Gulf of Maine, a rapidly changing region of the world's oceans. A comparison of the oxygen isotopes of the water (δ¹⁸O_water) and salinity data reveal that water samples fall on a tight, linear mixing line between fresher shelf water and saltier slope waters, with the freshwater endmember originating from much higher latitudes (the Gulf of St. Lawrence and the Labrador Sea). Some subtle differences in freshwater endmembers are observed between the three different deep basins in the Gulf of Maine. These differences are likely reflecting differences in freshwater input and vertical mixing between the different basins. Additionally, these water samples have lower δ¹⁸O_water values for a given salinity value than previously published values of marine water mass endmembers. This offset may be related to systematic changesin water mass endmember values or year to year variability, as well as differences in the proportions of water masses entering the Gulf of Maine. Nitrogen and oxygen isotopes of dissolved nitrate (NO₃⁻; δ¹⁵N_NO3- and δ¹⁸O_NO3-, respectively) measured in the water samples suggest a strong influence of phytoplankton assimilation near the surface in both isotopic systems. Combining these two datasets using Δ(15, 18) to look at the rates of fractionation between the two isotope systems reveals potential water column nitrification above 100 m in most places in the Gulf of Maine. This finding provides support for previous hypotheses of water column nitrification in the Gulf of Maine based on nutrient distribution and nitrogen box modeling. However, these calculations rely on the assumption that all nitrate is sourced from deeper waters. It is possible these results are instead caused by NO₃⁻ from different sources at the surface and therefore do not necessarily indicate the presence of nitrification.

Introduction

The Gulf of Maine is changing significantly. This area has been warming rapidly in the last several decades [Pershing et al., 2015; Thomas et al., 2017], due to both surface warming as well as changes in ocean currents, likely as the result of a melting Arctic [Pettigrew et al., 2008; Pettigrew et al., 2011; Smith et al., 2012]. Ocean acidification also poses a severe threat to the many fisheries in the region [Gledhill et al., 2015], as does decreased dissolved oxygen content [Bopp et al., 2013; Petrie and Yeats, 2000]. Understanding how Gulf of Maine ecosystems will be impacted by continued changes requires an in-depth understanding of how the many components of the Gulf of Maine ecosystem interact and behave, both today and in the past. Two components of these ecosystem interactions are water mass origin and nitrogen cycling processes. This study seeks to better understand these two components of the modern day Gulf of Maine hydrographic system using stable isotopes measured in water samples taken from throughout the Gulf of Maine region.

The Gulf of Maine is a semi-enclosed sea on the east coast of North America and a highly productive region of the world's oceans [Fig. 1; Townsend et al., 2015]. The Gulf of Maine is fed primarily by nutrient rich slope waters, which enter at depth through the Northeast Channel and mix together during transport in the Northeast Channel and the interior Gulf of Maine [Bigelow, 1927; Ramp et al., 1985; Smith et al., 2001], and Scotian Shelf Water (SSW), entering at the surface around Cape Sable, Nova Scotia [Gatien, 1976]. Deep slope waters are composed of Warm Slope Water (WSW), which forms adjacent to the Gulf Stream and is a mixture of Gulf Stream waters, shelf waters, and North Atlantic Central Water [Gatien, 1976; Townsend et al., 2015] and Labrador Slope Water (LSW), which originates in the Labrador Sea and is transported southward via the Labrador Current [Fig. 1; Gatien, 1976]. SSW is composed of LSW, St. Lawrence River Water and Labrador Shelf Water. Surface circulation in the Gulf of Maine is largely cyclonic, with coastal circulation primarily consisting of the Eastern Maine Coastal Current, which turns offshore near the mouth of the Penobscot River in the summer season, and the Western Maine Coastal Current [Pettigrew et al., 2005]. The Gulf of Maine has three deep (270–370 m deep) ocean basins, including Georges Basin, near the Northeast Channel entrance to the Gulf of Maine, Jordan Basin to the east, and Wilkinson Basin to the west (Fig. 2). Deep slope waters first enter Georges Basin before flowing into the other two basins. Slope waters and surface waters are mixed through various physical mechanisms, including storm mixing and winter vertical convection [Hopkins and Garfield, 1979]. Deep and bottom waters are also formed in the winter in Wilkinson Basin due to winter overturning, However, waters in Georges and Jordan Basin are too saline and therefore dense due to significant slope water concentrations to allow for bottom water formation in the winter. Winter overturning generates winter water in the upper water column in Georges and Jordan Basin, which is then capped by insolation and spring freshwater runoff in the spring and summer, resulting in Maine Intermediate Water [Hopkins and Garfield, 1979].

There have been several studies using oxygen isotopes of water (δ¹⁸O_water) to look at water sources and water mass mixing in the Gulf of Maine and surrounding region [Chapman et al., 1986; Chapman and Beardsley, 1989; Fairbanks, 1982; Houghton and Fairbanks, 2001; Khatiwala et al., 1999; Whitney et al., 2017]. δ¹⁸O_water can be used as a water mass tracer because its value is dependent on the latitude of origin [Craig, 1961], as well as precipitation and evaporation, and is not altered by biological processes. δ¹⁸O_water studies in the Gulf of Maine suggest freshwaters in the interior of the Gulf of Maine originate from high latitude locations, except for very coastal locations, where waters are heavily influenced by Maine River Water [MRW; Whitney et al., 2017]. The δ¹⁸O_water studies, while informative, suffer from a lack of comprehensive spatial coverage as most of these studies include only one or two transects that penetrate only part way into the Gulf of Maine [Chapman et al., 1986; Houghton and Fairbanks, 2001; Khatiwala et al., 1999] while Fairbanks [1982] only present data from one location (Wilkinson Basin) and Whitney et al., [2017] present data only from coastal locations. The addition of the data presented in this study, which is taken from 44 different locations throughout the Gulf of Maine (Fig. 2) over a nine-day period, allows both for a more spatially comprehensive survey of Gulf of Maine δ¹⁸O values as well as an analysis of the stability of source water δ¹⁸O_water through recent times.

Water mass origins in the Gulf of Maine have also been assessed in terms of nutrient concentrations and in particular nitrogen abundance and nitrogen cycling processes. The majority of “new” nutrients in the Gulf of Maine enter the Gulf of Maine in slope waters through the Northeast Channel. WSW has relatively high nutrient concentrations, including nitrate (NO₃⁻) concentrations of >23 μM, compared to those of LSW, which are closer to 16–17 μM [Townsend and Ellis, 2010; Townsend et al., 2014]. Strong tidal mixing close to the Bay of Fundy brings these cold, high-nutrient waters to the surface [Townsend et al., 1987]. These waters are then carried to the western Gulf of Maine via the coastal current and provide much of the nutrients for the biological productivity in the area. Estimates of slope water nutrients that are upwelled to the surface range between 23% [Townsend, 1998] and 44% [Townsend et al., 1987], with lower estimates suggesting that SSW is equally as important as slope water in providing nutrients for primary production in surface waters [Townsend, 1998].

There have been several nitrogen budgets suggested for the Gulf of Maine over the last several decades [Christensen et al., 1996; Schlitz and Cohen, 1984; Townsend, 1991; Townsend, 1998]. The most thorough of these nitrogen budgets was performed by Townsend [1998], who used a box model to show nitrogen cycling in the Gulf of Maine and suggest that NO₃⁻ must in part be coming from internal water column nitrification. Townsend showed that the flux of nitrogen from external sources (WSW, LSW, SSW, MRW, and atmospheric deposition) to Gulf of Maine surface waters was only enough to support 59 gC m⁻² yr⁻¹ of new primary productivity, 20% of the estimated total primary production in the Gulf of Maine. The ratio of this NO₃⁻ to recycled ammonia is much lower than would be expected for a productive region such as the Gulf of Maine. These results therefore suggest that a source of NO₃⁻ is missing from the box model. Townsend suggests that this source is water column nitrification below the surface layer but notes that such nitrification must be investigated in the water column. Benitez-Nelson et al., [2000] supported this hypothesis by using the radionuclide ⁷Be and NO₃⁻ plus nitrite (NO₂^-) profiles to show that vertical eddy diffusion in the western Gulf of Maine, which brings nutrients including NO₃⁻ formed from nitrification into the surface waters, was enough to support the export flux of particulate organic carbon (i.e. biomass from primary production) the authors calculated for the western Gulf of Maine using ²³⁴Th. To our knowledge, further field studies of water column nitrification processes in the Gulf of Maine have not been performed.

Here we present isotopic data from water samples collected during a nine-day research cruise in the Gulf of Maine in October 2016. These data offer a snapshot of water conditions towards the end of the productive season in the Gulf of Maine. The goal of this study was to investigate Gulf of Maine water properties through isotopic analyses. To achieve this goal, we had two main objectives: Investigate (1) water mass source and mixing and (2) nitrogen cycling within the Gulf of Maine. To accomplish objective 1, we analyzed collected water samples for δ¹⁸Ο_water. To accomplish objective 2, we analyzed collected water samples for nitrogen isotopes of dissolved nitrate (δ¹⁵N_NO3-) along with oxygen isotopes of dissolved nitrate (δ¹⁸O_NO3-), and the variable Δ(15, 18), defined below in Section 2 and Equation (1), that can be used to compare differences in changes between δ¹⁵N_NO3- and δ¹⁸O_NO3- throughout the water column. Taken together, δ¹⁵N_NO3-, δ¹⁸O_NO3-, and Δ(15, 18) can be used to better understand nitrogen cycling and utilization processes in the water column, particularly when using the Δ(15, 18) as different nitrogen cycling processes influence δ¹⁵N_NO3- and δ¹⁸O_NO3- in unique ways (described in detail in Section 4.3). Accomplishing the objectives stated here provides necessary information in order to better understand water properties and characteristics of the Gulf of Maine basins and surrounding shelf. Such information is critical in order to establish modern water column property baselines with which to compare both future expected changes in the Gulf of Maine region as well as past changes recorded in marine environmental proxies [e.g. Keigwin and Pilskaln, 2015; Wanamaker et al., 2008; Whitney et al., 2019].