Ocean mixing in deep-sea trenches: New insights from the Challenger Deep, Mariana Trench

Highlights

• Improved deep CTD observations from Challenger Deep, Mariana Trench.

• Turbulence estimates show values comparable to Puerto Rico Trench values.

• Turbulence generation via inertio-gravity waves.

• No crucial role suggested for seafloor topography in deep trench mixing.

• High-resolution multibeam echosounder gives new deepest point on Earth estimate.

Abstract

Reliable very deep shipborne SBE 911plus Conductivity Temperature Depth (CTD) data to within 60 m from the bottom and Kongsberg EM122 0.5° × 1° multibeam echosounder data are collected in the Challenger Deep, Mariana Trench. A new position and depth are given for the deepest point in the world's ocean. The data provide insight into the interplay between topography and internal waves in the ocean that lead to mixing of the lowermost water masses on Earth. Below 5000 m, the vertical density stratification is weak, with a minimum buoyancy frequency N = 1.0 ± 0.6 cpd, cycles per day, between 6500 and 8500 m. In that depth range, the average turbulence is coarsely estimated from Thorpe-overturning scales, with limited statistics to be ten times higher than the mean values of dissipation rate ε_T = 3 ± 2 × 10^–11 m² s⁻³ and eddy diffusivity K_zT = 2 ± 1.5 × 10⁻⁴ m² s⁻¹ estimated for the depth range between 10,300 and 10,850 m, where N = 2.5 ± 0.6 cpd. Inertial and meridionally directed tidal inertio-gravity waves can propagate between the differently stratified layers. These waves are suggested to be responsible for the observed turbulence. The turbulence values are similar to those recently estimated from CTD and moored observations in the Puerto Rico Trench. Yet, in contrast to the Puerto Rico Trench, seafloor morphology in the Mariana Trench shows up to 500 m-high fault scarps on the incoming tectonic plate and a very narrow trench, suggesting that seafloor topography does not play a crucial role for mixing.

Introduction

Life exists at great ocean depths in the ocean's hadal zone to water depths of over 10,000 m in deep sea trenches like the Mariana Trench (Jamieson, 2015, Gallo et al., 2015, Nunoura et al., 2015). As the deepest life requires sufficient supply of nutrients and energy in form of chemical species, the ocean, even at these great depths, has to be in motion and cannot be stagnant. In analogy with the atmosphere, where breathing by inhaling of oxygen would be impossible without turbulent motions, life in the hadal zone requires turbulent rather than laminar flows for survival. Mainly due to the logistical problems imposed by the large hydrostatic pressure which normal oceanographic equipment does not withstand, little is known about the physical oceanography of deep trenches and nothing about the physics that govern the turbulent processes. For example, turbulence microstructure profiles do not go deeper than 6000 m to date. As an indicator for upper trench turbulence, recent yearlong high-resolution temperature measurements from around about 6000 m just below the ‘top’ of the Puerto Rico Trench suggest turbulence generation by the interaction of large-scale 20–100 days periodic boundary currents with near-inertial and tidal internal wave breaking (van Haren and Gostiaux, 2016). Turbulence estimates from these data correspond to estimates from shipborne Conductivity Temperature Depth CTD data averaged over a suitable depth range of 600 m and the overturn shapes suggest shear-convective turbulence (van Haren, 2015). These shear-convective turbulent mixing processes found in trenches are quite different in magnitude from the mainly shear-driven turbulence found in deep passages through ridges and between islands (e.g., Polzin et al., 1996; Lukas et al., 2001; Alford et al., 2011). However, in both cases turbulence is inherently pulse- and intermittent-like with overturn sizes reaching 200 m.

The only moored and hourly sampled measurements so far, by Taira et al. (2004), near the deepest point on Earth, the bottom of the Challenger Deep--Mariana Trench, showed typical current speeds of 0.04 m s⁻¹ with a dominant semidiurnal tidal periodicity. Although Taira et al. (2004) did not show internal wave band spectra they mentioned sub-peaks at diurnal and inertial frequencies. These data already suggested that waters are not stagnant. The observed semidiurnal currents may be related to internal tides, whether propagating from remote source Luzon Strait (Morozov, 1995) or from local source Mariana Arc (Jayne and St. Laurent, 2001).

Water characteristics are also barely known near the bottom of the Challenger Deep (Table 1). This is mainly because few oceanographic research vessels are equipped with cables that are more than 11 km long. As an alternative and following discrete inverse thermometer readings from R/V Vityaz in the late 1950's, a small free-falling water-sampling device equipped with reversing thermometers was dropped to the bottom of the Challenger Deep in 1976 (Mantyla and Reid, 1978). These data are also used by Taira et al. (2005) as a reference for the first deep CTD-cast attached to a custom-made titanium wire in the Challenger Deep, down to 10,877 m. They used a SeaBird Electronics SBE-911 CTD at a station that was 40 km east of the site where Mantyla and Reid (1978) deployed their water sampler. Manned (Gallo et al., 2015) and un-manned (Nunoura et al., 2015) submarines carried CTDs (an SBE-49 in the latter case), but these data have not been analyzed and published for detailed water characteristics.

In this paper, we report on new high-resolution SBE-911 CTD casts into the Challenger Deep, reaching a depth of 10,851 m in 10,907 m water depth at a location about 2 km east of Mantyla and Reid's position. The CTD data deliver T, Salinity S and stratification information and thus information on the internal wave band below 5000 m of the surrounding ocean floor and near the maximum capability of standard oceanographic instrumentation. A second objective of this experiment is to gain a first impression of the internal wave-induced turbulence variation with depth down to near the bottom of the Challenger Deep. As a third objective, we use the simultaneously acquired high-precision multibeam echosounder data that were calibrated with the local CTD data to validate the most recent estimates of Challenger Deep's greatest depth (Gardner et al., 2014).