WHOI  WHOI People  

Tracing glacial meltwater in Sermilik Fjord, Greenland


I am requesting funding to make oxygen isotope measurements on water samples from Sermilik Fjord, a major glacial fjord in Greenland. These measurements, combined with temperature and salinity data, would allow me to perform an improved water property analysis to track glacially modified waters and infer export pathways of glacial meltwater.


Many recent studies indicate that the ocean plays a key role in the mass-balance of the Greenland Ice Sheet (GIS). Glaciers drain the ice sheet into deep, strongly stratified fjords, which contain a subsurface layer of Atlantic-origin water. A recent acceleration, thinning and retreat of many outlet glaciers has been linked to warming ocean waters off the coast of Greenland, suggesting that enhanced submarine melting triggered glacier changes. A dearth of observations from within these fjords, however, has limited our understanding of basic fjord circulation, water property variability near the glaciers and controls on submarine melting. The heat and freshwater transport, likely controlled by various modes of circulation, must be resolved in order to predict future changes to the ice sheet and global sea level.

Thesis work

I am studying the intersection of external forcing with glacial freshwater forcing in a major fjord to develop a dynamical framework for assessing circulation and heat transport. My primarily study site is Sermilik Fjord in southeast Greenland. Helheim glacier, the third largest outlet of the GIS drains into Sermilik, depositing freshwater as submarine meltwater, subglacial discharge and icebergs. The primary components of my thesis include: 1. Examine externally-driven components of the circulation (e.g. local and remote wind-forcing, tides, etc.) which dominate the velocity field and are important for flushing the fjord with new shelf waters. 2. Examine the glacier-driven circulation from inputs of freshwater. This signal is largely masked in velocity records by the more energetic externally-driven flows. Water property analysis can be used to track glacially modified waters and infer export pathways of submarine meltwater. 3. Study the intersection of these two regimes with observations and a model. Ultimately, I want to understand their relative contributions to heat transport towards the glacier.

Tracing glacial meltwater

This proposal pertains to the second step of my thesis – tracking glacial meltwater in the fjord to infer export pathways – which has been partially completed. In typical estuaries, freshwater is found at the surface and can be identified by measuring salinity alone. In large glacial fjords, however, freshwater enters at depth and flows into a stratified ambient; meltwater plumes will rise until they have entrained enough ambient water to reach a neutral density, which can be hundreds of meters below the surface. Building on previous work (Straneo et al., 2011), I have worked with hydrographic fjord surveys of temperature and salinity to distinguish shelf water masses from glacial freshwater inputs. In the winter, this task is tractable as there are only three significant water masses in the fjord: Polar Water (PW) and Atlantic Water (AW) from the shelf and submarine meltwater from the glacier. Effective potential temperature1 (θe) and salinity (S) are conservative tracers (away from the surface) and, if we know the end-member properties of each water mass and follow the method of Jenkins et al. (1999), we can directly calculate the meltwater fraction at each point. This analysis shows that meltwater exits at the surface and at the mid-depth stratification peak at the PW/AW interface, in agreement with modeling studies (e.g. Xu et al., 2012; Sciascia et al., 2013).

In the summer, a fourth water mass – subglacial discharge – is present. As air temperatures rise above freezing, surface melt drains through the ice sheet and enters the fjord at the base of the glacier’s terminus. Subglacial discharge and submarine melt are both freshwater inputs (S=0) but have different θe signatures2 and thus must be treated as different water masses. From summer surveys of θe and S, I have been able infer regions of glacial modification, but the analysis is limited; with four water masses and two tracers, the system is underdetermined. A third tracer is needed to distinguish between water masses and to quantitatively assess meltwater fractions.

δ18O as a 3rd tracer

I hope to add oxygen isotope measurements to my water property analysis. The ratio of stable oxygen isotopes3 18O) is a good tracer for glacial meltwater because glacial ice is highly depleted in the heavy isotope compared to ocean water masses. A literature review suggests that the four water masses of interest should have distinct end-member values4, which we can confirm with samples of AW and PW outside the fjord (which are unmodified by the glacier) and ice samples. We expect δ18O to mix as a conservative tracer within the fjord. Therefore, measurements of δ18O, S and θ should allow me to distinguish between water masses and examine the spread of submarine meltwater and subglacial discharge through the fjord.


I have already obtained water samples from Sermilik Fjord. While performing a CTD survey of the fjord in August, I used Nisken bottles to collect water at depth ranges of 10-540m in the upper fjord and 1-585m on the shelf outside the fjord. Additionally, I collected two water samples from icebergs in the upper fjord. This fieldwork was covered by an NSF extension to recover lost moorings, so there is no funding allocated for water analysis. If funded by the COI, I would send these water samples to the UC Davis Stable Isotope Facility to be run on their Laser Water Isotope Analyzer. I would continue to work with chemists5 who are experts on isotopic measurements throughout the project as I incorporate this data into my study of the heat and freshwater transport in glacial fjords.


Azetsu-Scott, K. & Tan, F., 1997. Oxygen isotope studies from Iceland to an East Greenland Fjord: Behaviour of glacial meltwater plume. In Marine Chemistry. pp. 239–251.

Gade, H.G., 1979. Melting of ice in sea water: A primitive model with application to the Antarctic ice shelf and icebergs. Journal of Physical Oceanography, 9(1), pp.189–198.

Jenkins, A., 1999. The impact of melting ice on ocean waters. Journal of Physical Oceanography, 29(9), pp.2370–2381.

Reeh, N. and H.H. Thomsen. 1993. Using stable isotopes as natural tracers to delineate hydrological drainage basins on the Greenland ice-sheet margin. Chemical Geol., 109(1–4), 281–291.

Straneo, F. et al., 2011. Impact of fjord dynamics and glacial runoff on the circulation near Helheim Glacier. Nature Geoscience, 4(5), pp.322–327.

Sutherland, D.A. et al., 2009. Freshwater composition of the waters off southeast Greenland and their link to the Arctic Ocean. Journal of Geophysical Research, 114(C5), p.C05020

When ice melts into water, the latent heat to melt ice can be accounted for by using effective potential temperature, θe=θf – L/c_p – ci/cp(θfθi), where θf is the freezing temperature, θi the ice temperature, !!the latent heat of fusion, and cp and ci the heat capacities of water and ice, respectively (Gade, 1979).

2 Submarine meltwater has end-member properties of S =0 and θe = -90°C, whereas subglacial discharge has already melted before entering fjord, so its end-member values are S =0 and θe = 0°C.

3 The oxygen isotopic ratio is defined as δ18O = [(18O/16O)obs / (18O/16O)eq – 1] × 1000.

4 Water mass end-member values of δ18O: submarine melt = –28 ± 2 ‰; subglacial discharge = –20 ± 5 ‰; PW = –1 ± 0.1 ‰; AW = 0.3 ± 0.1 ‰(Reeh & Thompsen, 1993; Azetsu-Scott & Tan, 1996; Sutherland et al., 2009).

5 I have already discussed this project with several scientist who are familiar with oxygen isotope measurements,including: Brice Loose at URI; Alison Criscitiello, Britta Voss & Bernhard Peucker-Ehrenbrink at WHOI.

Last updated: December 17, 2013