Description of the data contained in this collection.
   This data set contains ocean profile data from four geographic regions originally acquired by the Ocean Observatories Initiative (see https://oceanobservatories.org/).  The sites are: Argentine Basin (42? 58.9’ S, 42? 29.9’ W, water depth 5200 m), Southern Ocean (54? 28.1’ S, 89? 22.1’ W, 4800 m), Station Papa (50? 4.2’ N, 144? 47.9 W, 4219 m), and the Irminger Sea (59? 58.5’ N, 39? 28.9’ W, 2800 m).  Upon the OOI program inauguration in fall 2014 – spring 2015 (Station Papa in summer 2013), annually-fielded arrays of four moorings were installed at each of their “global” sites.  The ~triangular arrays consisted of a surface mooring supporting atmospheric boundary layer sensors and discrete CTDs, ADCPs and point current meters in the upper ocean, an adjacent subsurface mooring some 4-8 km distant supporting one or two McLane Moored Profiler (MMP) vehicles fitted with a CTD and point current meter, and two so-called subsurface flanking moorings 40-60 km away with additional discrete sensors.  (The Station Papa surface mooring is maintained by the NOAA Pacific Marine Environmental Laboratory.)  The two southern hemisphere sites were discontinued in early 2018; the other two sites are still operational as of this writing.    
   Research initiatives were developed using these data that required observations spanning the full water column.  An added complication is that no single OOI mooring sampled velocity, temperature and salinity over full depth, thus such profiles had to be synthesized by combining observations from multiple moorings and conducting vertical interpolation/extrapolation.  
   The majority of global site OOI instrumentation is deployed in the upper few hundred meters of the ocean; the MMPs provided observations over the balance of the water column.  The OOI operators programmed the profilers to begin one-way traverses every 20 hours during their deployments, with each traverse taking two to three hours.  Owing to the great ocean depths at the Argentine Basin, Southern Ocean and Station Papa sites, to achieve year-long endurance, two MMPs were fitted on the moorings with profile start times offset by four hours.  Using procedures developed by the PIs, the raw MMP data were processed to ˝ - dbar-resolution profiles of horizontal velocity in earth coordinates, temperature, salinity and time (the latter to account for the MMP finite travel speed and the staggered start times of the paired MMPs for the analysis of near-inertial and semi-diurnal variability).  Ship-based, calibrated hydrographic data from the sites were used to derive profile-to-profile adjustments of the MMP conductivity data to achieve consistency in deep water potential temperature-salinity relationships, with upward extrapolation across the depth gap between the paired MMPs to reference the shallower profiler’s CTD sensor.  
   The OOI MMP data were extended to the surface using the available ADCP and fixed-depth CTD sensors on the adjacent moorings.  The closest hourly ADCP estimates at the times of the shallowest MMP observations were linearly interpolated to ˝ dbar resolution and appended atop the MMP profiles.  Spurious estimates are noted in the upper ~50 m of the water column (likely contamination from side lobe surface reflections).  Similarly, the discrete CTD temperature and salinity estimates from adjacent moorings at the time of each MMP profile were extracted, inserted at their respective depth bins and also interpolated to the same vertical resolution.  In some cases, MMPs did not fully span their programmed depth interval, and upper-ocean sensor data were not always available.  In these cases, the individual profiles were filled in by vertical extrapolation and/or interpolation.  
   Using the synthesized full-depth temperature and salinity data, profiles of the vertical displacement of neutral density surfaces were constructed with reference to annual- or seasonal-mean density profiles.  The reference profiles, formed by averaging ensembles of synoptic neutral density profiles on pressure surfaces, were extended vertically by linear extrapolation, up and down, to span the observed range of neutral density at each site.  (These inferred vertical displacements are unphysical in the upper ocean where air-sea exchange induced density changes that were not associated with vertical heave, and inaccurate or unphysical at the bottom where density variability was the result of across-isobath bottom flow and/or water mass variability.)  Displacements were estimated in depth space using a latitude-dependent conversion from pressure in decibars ignoring the contributions from dynamic height.