<rahmann et al. 
periods were additionally equipped with biogeochemical sensors 
recording variables such as partial pressure of CO2 (pCO»2), 
Auorescence, and particle fluxes. Instruments with pressure, 
temperature, conductivity, and oxygen sensors were calibrated 
in situ immediately prior to and after a mooring deployment 
period by attaching them to the CTD frame during CTDO 
casts. Correction terms were then developed from the difference 
between the sensor readings and the calibrated CTDO data 
during several minute-long calibration stops. These correction 
terms were then applied to the full deployment periods. This 
ensured best data quality with recognition of potential sensor 
drifts and also allowed for the estimation of calibration and 
measurement errors (Hahn et al., 2014; Bittig et al., 2018; Berx 
et al., 2019). Moored Acoustic Doppler Current Profiler bin 
depths were corrected using the sound speed at instrument depth 
following the approach by Shcherbina et al. (2005). Velocities 
were not corrected, but respective measurement errors were 
assumed as described in Hahn et al. (2014). For all instruments 
within a mooring that did not record pressure, full deployment 
pressure time series were estimated by linearly interpolating 
between the instruments having a pressure sensor. 
Salinometry 
The conductivity sensors of the CTD were calibrated against 
international Association for the Physical Sciences of the Ocean 
(IAPSO) Standard Seawater samples with known conductivities 
using Guildline Autosal B instruments. On all cruises two 
Autosals were available and used to measure between 100 and 
1000 samples (typically 300-400 for a cruise or 4-5 per CTD 
cast). The procedures used for the calibration followed the 
recommendations in the GO-SHIP manual (Kawano, 2010). The 
results from the salinometer measurements are included in the 
source files for CTD data published on PANGAEA (Krahmann 
and Mehrtens. 20210). 
Autonomous Gliders 
Autonomous gliders were deployed during several cruises but 
also as stand-alone missions independent from large research 
vessels (Krahmann and Mehrtens, 2021a; see Table 2 and 
Supplementary Table 4). Two different generations of Teledyne 
Webb Research Slocum gliders were used, G1 and G2. All gliders 
were equipped with Seabird CTD systems, G1 gliders with an 
unpumped and G2 gliders with a pumped version, respectively. 
An Aanderaa optode was present on all gliders to observe 
dissolved oxygen concentrations. Optical fluorescence and 
backscatter sensors manufactured by Wetlabs were also present 
on all gliders albeit in different configurations. They allowed the 
determination of CHL-a (excitation and emission wavelengths 
of 470 and 695 nm, respectively) and CDOM (excitation 
and emission wavelengths of 370 and 460 nm, respectively) 
concentrations and the turbidity (scattering wavelength of 
470 nm) of the waters. All glider data were processed using a 
GEOMAR-developed software (Thomsen et al., 2016) resulting in 
gridded fields for all observed variables. During a small number 
of glider deployments, a Seabird/Satlantic SUNA Nitrate sensor 
was attached to a glider. SUNA data were processed following the 
procedures outlined in Sakamoto et al. (2009) and Sakamoto et al. 
-rontiers in Marine Science | www.frontiersin.orc 
SFB754 Data Legacy 
(2017) and calibrated against Nitrate measurements from nearby 
CTDO casts with discrete Nitrate measurements. Microstructure 
sensors were also attached to gliders on several deployments (see 
following section). 
Ocean Turbulence Measurements 
Ocean turbulence measurement programs were carried out 
during 22 cruises to quantify the dissipation rate of turbulent 
kinetic energy and infer rates of turbulent mixing (Dengler 
and Mehrtens, 2021; see Table 2 and Supplementary Table 5). 
The shipboard microstructure profiling systems (MSS) were 
manufactured by Sea & Sun Technology and consisted of a 
profiler (MSS90-D, S/N 26, 32, and 73), a winch having 500- 
L000 m of cable, and a data interface. All profilers were 
2quipped with three microstructure shear sensors, a fast-response 
(emperature sensor (PF07), an acceleration sensor, and two tilt 
sensors as well as conductivity (Sea & Sun Tech.), temperature 
(Sea & Sun Tech.), pressure (Keller), turbidity (Seapoint), and 
oxygen sensors sampling with a lower response time. The 
profilers were optimized to sink at a rate of 0.5-0.6 m s7}. 
Standard processing procedures were used to determine the rate 
of kinetic energy dissipation of turbulence in the water column 
{see Schafstall et al., 2010). 
Additionally, during several autonomous glider missions, a 
microstructure probe was mounted to the top of the gliders. 
These probes (MicroRider) were manufactured by Rockland 
Scientific and carried two microstructure shear and temperature 
sensors as well as pressure, accelerometer and tilt sensors. The 
data processing is detailed in Foltz et al. (2020). 
Shipboard Acoustic Doppler Current Profiler (SADCP) 
Measurements 
Shipboard acoustic doppler current profiler (SADCP) data were 
acquired on 33 of the research cruises (Krahmann and Mehrtens, 
2021e; see Table 2 and Supplementary Table 6). On FS Meteor, 
FS Maria S. Merian, and FS Sonne II two Teledyne RDI Ocean 
Surveyor systems with 38 and 75 kHz transmission frequency 
were used, while on NO /’Atalante a single 75 kHz system was 
used. All data were processed with a software package developed 
at GEOMAR following the GO-SHIP standards (Firing and 
Hummon, 2010). The data were subsequently averaged over 
L-min intervals, converted to a NetCDF based format, and 
published. For a small number of cruises, the signal strength 
information of the SADCP data has been used to estimate the 
backscatter in the ocean. These data sets were processed following 
Mullison (2017) and published separately from the regular 
SADCP data (Krahmann and Mehrtens, 2021e; see Table 2 and 
Supplementary Table 6). 
Underway Conductivity-Temperature-Depth (UCTD) 
and Rapidcast Measurements 
During the second funding phase (2012-2015) a new CTD 
system became available that could be deployed from a moving 
ship. First a Teledyne Oceanscience UCTD and later a Teledyne 
Oceanscience Rapidcast system were acquired and deployed 
successfully on several cruises (Krahmann and Mehrtens, 2021g; 
see Table 2 and Supplementary Table 7). They allowed for the 
3eptember 2021 | Volume 8 | Article 72330«