Mapping Internal Tides From Satellite Altimetry Without Blind Directions

Abstract

Satellite altimetry is one practical technique for observing internal tides on the global scale.However, it is a great challenge to extract weak internal tide signals. This paper presents a new technique for mapping internal tides from satellite altimeter data. Along-track high-pass filtering is needed to remove long-wavelength nontidal noise and the barotropic tidal residual; however, the filter also removes internal tides having large angles with respect to satellite ground tracks. It thus causes blind directions in mapping internal tides from satellite altimetry: Generally west-east propagating internal tides are missed. The new technique addresses the blind-direction issue by replacing the problematic one-dimensional (1-D) high-pass filter with a two-dimensional (2-D) band-pass filter. This mapping technique is able to retrieve ubiquitous westbound and eastbound internal tides not captured in previous estimates. Long-range westbound and eastbound waves travel over thousands of km from numerous generation sites such as the Emperor seamount chain, the Hawaiian Ridge, and the Kermadec trench. Evaluation using independent Cryosat-2 data reveals that the new internal tide model may reduce more SSH variance than a model built in 2016 does in regions of strong internal tides. However, this mapping technique makes no improvement in strong boundary current regions, due to the dominance of mesoscale motions. Moreover, the new internal tide model contains leaked noise from westward propagating tropical instability waves (TIWs), which can be suppressed by prior along-track high-pass filtering. This paper suggests that better internal tide models may be constructed using both 1-D and 2-D filters with optimized parameters. Plain Language SummaryInternal tides are ubiquitous in the ocean and play an important role in various ocean processes. It is a significant task in the community to construct internal tide models. Satellite altimetry is the most important technique for observing global internal tides. However, it is a great challenge to extract weak internal tides, mainly due to the low spatiotemporal sampling rate of satellite altimetry. Fortunately, dedicated mapping techniques make the best use of satellite altimeter data to construct better and better internal tide models. In this paper, I make improvements by addressing an issue in previous techniques. Along-track high-pass filtering is widely used in previous studies to remove long-wavelength nontidal noise and barotropic tidal residual. Unfortunately, the filter also removes internal tides having large angles with respect to satellite ground tracks and misses west-east propagating internal tidal waves. The new internal tide model developed by the new technique improves over previous models in retrieving westbound and eastbound internal tides. The results are confirmed using independent Cryosat-2 data. The new model shows that west-east propagating internal tides occur throughout the ocean and can travel up to 3,800 km from Hawaii to California.

Southward Internal Tides in the Northeastern South China Sea

“…This technique has been improved by combing with 2‐D band‐pass filtering (Zhao, 2019). The latter is key to further filter out nontidal noise, so that even weaker internal tides can be observed.…”

“…The present paper shows southward internal tides from the continental slope to the south of the Taiwan Strait by a new mapping technique. The technique combines plane wave analysis and two‐dimensional (2‐D) band‐pass filtering (Zhao, 2019): The former can resolve waves of different propagation directions; the latter can significantly suppress nontidal noise. The technique makes it possible to extract the much weaker internal tides previously neglected.…”

“…The data have been corrected for atmospheric effects, surface wave biases, and geophysical effects in the same manner (section 2.1). The CryoSat‐2 data have been previously used for evaluating empirical internal tide models developed using multisatellite data collected by exact‐repeat‐track altimeter missions (Ray & Zaron, 2016; Zaron, 2019; Zhao, 2019) and mapping M 2 internal tides (Zhao, 2016). In contrast, pointwise mooring measurements at individual stations are not well qualified for assessing empirical internal tide models, because moorings cannot resolve multidirectional waves and thus are subjected to complicated temporal and spatial variations.…”

Wind Wave Effects on Remote Sensing of Sea Surface Currents From SAR

“…Satellite remote sensing observations have provided a cost-effective means of monitoring ocean surface currents over the past two decades. Among present spaceborne remote sensing platforms, satellite altimeter is a baseline for the ocean current/circulation research, from the large scale to the submesoscale processes, including internal tides and internal waves, benefiting from the dynamical relationships between ocean currents and sea surface height (e.g., Morrow et al, 2019;Vergara et al, 2019;Zhao, 2019). For example, the geostrophic current retrieved from altimeter data has been widely used for ocean dynamics and climate change research (e.g., Dong et al, 2014).…”

Internal Tide Cycle and Topographic Scattering Over the North Mid‐Atlantic Ridge

“…In situ observations can only provide limited information and cannot be used to reconstruct the whole lifecycle at large scales. On the other hand, maps of the low-mode internal tide distribution can be estimated using long-duration satellite altimeter data (Zaron, 2019;Zhao, 2018Zhao, , 2019. However, they are restricted to the first two baroclinic modes and provide only rough estimates of the sources and sinks for the energy, with no further information on the physics behind.…”