Theses Doctoral

Moored observations of upper-ocean turbulence and polynya processes

Miller, Una Kim

The upper ocean mediates the transfer of heat and carbon between the atmosphere and ocean interior. The study of this dynamic environment, made possible in part by long-term time series gathered from oceanographic moorings, is therefore crucial to our understanding of Earth’s climate. In this thesis, we use moored datasets from the Southeast Pacific and Southern Oceans to explore two upper-ocean processes relevant to the transfer and eventual sequestration of atmospheric heat and carbon into the deep ocean: wind-, wave-, and buoyancy-forced turbulence and the release of brine in Antarctic polynyas that drives the formation of Antarctic Bottom Water (AABW).

In Chapter 1, we use measurements of turbulence kinetic energy (TKE) dissipation rate (ε) collected at 8.4 m depth on the long-established Stratus Mooring in the Southeast Pacific (20° S, 85° W) to assess the applicability of Monin-Obukhov similarity theory (MOST), Law of the Wall (LOW), and other boundary layer similarity scalings to turbulence in the upper ocean. TKE facilitates the mixing of heat, momentum, and solutes within and between the ocean and atmosphere and is generated in the upper ocean primarily by wind, waves, and buoyancy fluxes. Its production can generally be assumed to equal its dissipation, and measurements of ε therefore serve as a means for quantifying turbulence in a system. We present 9 months of ε measurements, a remarkably long time series made possible by the use of a moored pulse-coherent Acoustic Doppler Current Profiler (ADCP), a new methodology for measuring ε that uniquely allows for concurrent surface flux and wave measurements across an extensive length of time and range of conditions.

We find that turbulence regimes are quantified similarly using the classic Obukhov length scale (L_M=(u_*³)/(κ𝐵ₒ), where u_* is ocean-side friction velocity, κ is the von Kármán constant, and B_0 is surface buoyancy flux) and the newer Langmuir stability length scale (L_L=(〖u_s u〗_*²)/B_0 , where u_s is surface Stokes drift velocity), suggesting that u_* implicitly captures the influence of Langmuir turbulence at this site. This is consistent with the strong correlation observed between u_s and u_*, likely promoted by the steady southeast trade winds, and suggests that classic wind and buoyancy-based boundary layer scalings sufficiently describe turbulence in this this region.

Accordingly, we find the LOW (ε=(u_*³)/κz, where z is instrument depth) and surface buoyancy scaling (ε=B_0, where B_0 is destabilizing surface buoyancy flux) used in classic turbulence scaling studies, such as Lombardo and Gregg (1989), to describe our measurements well, and a newer scaling for Langmuir turbulence scaling based on u_s and u_* to scale ε well at times but to be overall less consistent than (u_*³)/κz. The performance of MOST relationships from prior studies in a variety of aquatic and atmospheric settings are also examined, and we find them to largely agree with our data in conditions where both convection and wind-driven current shear act as significant sources of TKE (-1<z/L_M <0). The apparent redundancy of Langmuir turbulence scaling and the sufficiency of LOW and MOST observed in this study may help inform the development of general circulation models (GCMs), which rely on boundary layer scaling to parametrize turbulent mixing in the upper ocean.

In Chapters 2 and 3, we focus on the Terra Nova Bay Polynya in the western Ross Sea of Antarctica, where High Salinity Shelf Water (HSSW) forms as a result of the cooling and salinification of the surface ocean by an intense katabatic wind regime and its associated ice production. HSSW is a precursor to AABW, a vital water mass that feeds the bottom limb of the meridional overturning circulation (MOC) and facilitates the sequestration of atmospheric heat and carbon into the abyss. A decades-long freshening trend in the salinity of Ross Sea HSSW resulting from increased glacial meltwater fluxes, and more recently, its abrupt reversal associated with the occurrence of a climate anomaly, have highlighted the complexity of this system and its sensitivity to changes in climate. Because the density of HSSW has a direct impact on the density of downstream AABW, and therefore the strength of the MOC, it is imperative to better understand the variability and mechanisms of HSSW formation. However, inhospitable wintertime conditions in this region severely restrict the collection of in-situ data in the presence of active brine rejection and HSSW formation.

Here, we present an unprecedented set of upper-ocean salinity, temperature, turbulence, current velocity, and acoustic surface tracking time series collected from a mooring in Terra Nova Bay during austral winter 2017. One poorly constrained aspect of HSSW in Terra Nova Bay is its rate of production, and in Chapter 2 we endeavor to produce the first production rate estimates to be based on in-situ salinity data. We find an average production rate of ~0.6 Sverdrups (10⁶ m³ s⁻¹), which allows us to improve on and validate an existing approach for estimating rates using parametrized net surface heat fluxes out of the polynya. We use this approach to examine interannual variability in production across the decade and find estimates of HSSW production in Terra Nova Bay to be largely increasing from 2015 onward. As higher production rates of Terra Nova Bay HSSW, the saltiest variety of HSSW across Antarctica, could increase the salinity of downstream AABW, this apparent increase may have played a previously unrecognized role in the recently observed recovery of AABW salinity in this region.

In Chapter 3, we examine a number of interconnected processes surrounding HSSW formation, including the coupling of salinity to winds, the breakdown of summer stratification that primes the water column for HSSW formation in the winter, wind-driven turbulence that facilitates the breakdown of stratification and mixing of HSSW to depth, and potential circulation pathways for HSSW formed at the mooring site. We find that salinity at the shallowest depth on the mooring line, 47 m, couples strongly to wind speeds measured at the nearby Automatic Weather Station (AWS) Manuela from April onward, demonstrating the dependence of polynya formation, ice production, and brine rejection on winds at the mooring site. Salinity at the deepest depth on the mooring line, 360 m, couples to salinity at 47 m beginning in June, following the progressive breakdown of lingering summertime water column stratification that previous studies have established as a prerequisite for HSSW formation in the winter.

We incorporate concepts from Chapter 1 to explore the scaling of turbulence in a polynya environment, finding that daily-averages of ε are sufficiently approximated according to the classic LOW scaling, despite visible evidence of Langmuir circulation in the polynya. To the best of our knowledge, this represents the first examination of turbulence scaling using in-situ time series measurements in an Antarctic polynya, an environment that connects the turbulent mixing of heat and solutes in the upper ocean to the properties of the deepest layer of the ocean. Lastly, we infer from current velocities and a late-winter coupling of salinity measured at our mooring to that measured by a second mooring within the Drygalski Basin that HSSW may travel one of two pathways following its formation at our mooring site: Directly southeastward into the Drygalski Basin or northeastward along with the cyclonic gyre of Terra Nova Bay. More mooring deployments across space and time within the bay are needed in order to further elucidate the variability and mechanisms surrounding HSSW formation, critical foci of study in the context of a rapidly changing Antarctic environment.


This item is currently under embargo. It will be available starting 2025-05-11.

More About This Work

Academic Units
Earth and Environmental Sciences
Thesis Advisors
Zappa, Christopher J.
Ph.D., Columbia University
Published Here
May 24, 2023