The authors present inferences of diapycnal diffusivity from a compilation of over 5200 microstructure profiles. As microstructure observations are sparse, these are supplemented with indirect measurements of mixing obtained from (i) Thorpe-scale overturns from moored profilers, a finescale parameterization applied to (ii) shipboard observations of upper-ocean shear, (iii) strain as measured by profiling floats, and (iv) shear and strain from full-depth lowered acoustic Doppler current profilers (LADCP) and CTD profiles. Vertical profiles of the turbulent dissipation rate are bottom enhanced over rough topography and abrupt, isolated ridges. The geography of depth-integrated dissipation rate shows spatial variability related to internal wave generation, suggesting one direct energy pathway to turbulence. The global-averaged diapycnal diffusivity below 1000-m depth is O(10 . The compiled microstructure observations sample a wide range of internal wave power inputs and topographic roughness, providing a dataset with which to estimate a representative global-averaged dissipation rate and diffusivity. However, there is strong regional variability in the ratio between local internal wave generation and local dissipation. In some regions, the depthintegrated dissipation rate is comparable to the estimated power input into the local internal wave field. In a few cases, more internal wave power is dissipated than locally generated, suggesting remote internal wave sources. However, at most locations the total power lost through turbulent dissipation is less than the input into the local internal wave field. This suggests dissipation elsewhere, such as continental margins.

Internal gravity waves, the subsurface analogue of the familiar surface gravity waves that break on beaches, are ubiquitous in the ocean. Because of their strong Internal gravity waves are propagating disturbances of the ocean's density stratification. Their physics resembles that of surface gravity waves but with buoyancy rather than gravity providing their restoring force -making them much larger (10's to 100's of meters instead of 1 to 10 meters) and slower (hours instead of seconds). Generated primarily by tidal flow past seafloor topography and winds blowing on the sea surface, and typically having multi-kilometer-scale horizontal wavelengths, their estimated 1 TW of deep-sea dissipation is understood to play a crucial role in the ocean's global redistribution of heat and momentum 12 . A major challenge is to improve understanding of internal wave generation, propagation, steepening and dissipation, so that the role of internal waves can be more accurately incorporated in climate models.The internal waves that originate from the Luzon Strait on the eastern margin of the South China Sea (SCS) are the largest documented in the global oceans ( Figure 1).As the waves propagate west from the Luzon Strait they steepen dramatically ( Figure 1a), producing distinctive solitary wave fronts evident in sun glint and synthetic aperture radar (SAR) images from satellites ( Figure 1b). When they shoal onto the continental slope to the west, the downward displacement of the ocean's layers associated with these solitary waves can exceed 250 m in 5 minutes 8 . On such a scale, these waves pose hazards for underwater navigation and offshore drilling 4 , and supply nutrients from the deep ocean that nourish coral reefs 1 and pilot whale populations that forage in their wakes 13 .Over the past decade a number of field studies have been conducted in the region; this work has been comprehensively reviewed 10,11 . All of these studies, however, focused on the propagation of the internal waves across the SCS and their interactions with the continental shelf of China. Until the present study there had been no substantial in situ data gathered at the generation site of the Luzon Strait, in large part because of the extremely challenging operating conditions. A consequence has been persistent 5 confusion regarding the nature of the generation mechanism 11 ; an underlying cause being the sensitivity of the models employed to the system parameters, such as the chosen transect for a two-dimensional model, the linear internal wave speed or the assumed location of the waves' origin within the Luzon Strait. Furthermore, the lack of in situ data from the Luzon Strait has meant an inability to test numerical predictions of energy budgets 9 and no knowledge of the impact of the Kuroshio on the emergence of internal solitary waves 11 .The goal of IWISE is to obtain the first comprehensive in situ data set from the Luzon Strait, which in combination with high-resolution three-dimensional numerical modeling supports a cradle-to-grave picture ...