Tuchen et al. 
ıntraseasonal variability in the central equatorial Atlantic is 
mainly attributed to TIWs in the upper ocean (Athie and 
Marin, 2008), while intraseasonal variability in the deep ocean 
is associated with the signature of equatorial Yanai waves (Ascani 
:t al., 2015; Tuchen et al., 2018, Körner et al., 2022). The 
observed and modelled interaction between intraseasonal 
equatorial waves and the aforementioned EDJs was found to 
maintain the deep equatorial circulation against dissipation 
‘Greatbatch et al., 2018; Bastin et al., 2020) pointing toward 
ihe importance of intraseasonal variability for equatorial 
ocean dynamics. 
These findings are largely based on, or underpinned by a 
unique and steadily expanding data set of current velocity 
observations in the central equatorial Atlantic Ocean. Since 
2001, current velocities have been measured almost continuously 
as part of a multilateral collaboration, the Prediction and Research 
Moored Array in the Tropical Atlantic (PIRATA), that regularly 
services a moored observatory located at 0°N/23°W (Bourlös et al., 
2019). The significance of this data set is characterized by the 
ıength of the time series and by the full-depth coverage of current 
velocity observations which allow for a detailed analysis of both 
upper-ocean and deep-ocean dynamics on a wide range of time 
scales and frequencies. For instance, it enables the decomposition 
of the current velocity time series into vertical modes pointing 
:oward the existence of resonant basın modes and identifying 
different sources of deep intraseasonal variability (Brandt et al., 
2016; Claus et al., 2016; Greatbatch et al., 2018; Tuchen et al., 2018; 
Körner et al., 2022). 
Here, we present 20 years of full-depth current velocity 
observations at 0°N/23°W. The aim of this study is to provide 
the scientific community with a publicly available reference data 
set that could be used in manifold ways, including, for instance, 
the validation of ocean models or reanalysis products. 
DATA AND METHODS 
Moored Observatory Design 
The 0°N/23°W moored observatory consists of a moored surface 
u0y, as part of the multilaterally-operated PIRATA array that 
was initiated in 1997 (Servain et al., 1998; Bourles et al., 2008; 
Bourles et al., 2019) and, within a few nautical miles of this 
mooring, a subsurface mooring that underwent several changes 
of design since its first deployment in 2001. To date, a total of 13 
mostly successive subsurface moorings have been deployed and 
zecovered between December 2001 and July 2021 (Table 1). 
From December 2001 to June 2006, current velocity 
measurements were acquired by the subsurface mooring as 
part of the French contribution to PIRATA. Since June 2006, 
after a total of three mooring periods (see Table 1), the existing 
subsurface mooring has been continued by GEOMAR. While the 
shallow acoustic Doppler current profiler (ADCP) remained a 
part of the PIRATA program, all other instrumentation has since 
vdeen provided by GEOMAR. Additional current velocity data 
were acquired at the PIRATA surface buoy to which a current 
meter is attached at a depth of 10 to 12 m since 2005, as well as a 
downward-looking ADCP between 2008 to 2009 located just 
Zrontiers in Marine Science | www frontiersin oru 
Moored Velocity Observations at 0°N, 23°W 
below the surface (Wenegrat et al., 2014; Wenegrat and 
McPhaden, 2015). The combined data set presented here is a 
combination of all available measurements from the subsurface 
mooring and the PIRATA moored surface buoy in close 
proximity to it. 
The general design of the subsurface mooring consists of 
current velocity measurements from moored ADCPs covering 
the upper 500 to 900 m of the water column and, since 2006, 
from a McLane moored profiler (MMP) covering depths 
between 1000 to 3500 m (since September 2016: 850 to 
3350 m). These two main components of the subsurface 
mooring are complemented by current meters in between the 
vertical range of the ADCP and the MMP or, since September 
2016, below the lower end of the MMP range. In addition, 
lowered ADCP (LADCP) profiles, carried out during the 
mooring service cruises, were included in the presented data set. 
In the following, we provide a brief description of these 
individual mooring components. A more detailed assessment 
of the processing methods of the individual instruments is 
beyond the scope of this data report. 
Moored ADCP Measurements 
Upper-ocean current velocity data were mainly acquired by 
moored ADCPs. From December 2001 to December 2002 an 
apward-looking 300-kHz ADCP was installed at about 150 m. 
After a gap of 14 months, the subsurface mooring was re- 
established in February 2004 with the deployment of two 
ADCPs: one upward-looking 300-kHz ADCP located at a 
nominal depth of about 90 m and one downward-looking 75- 
kHz ADCP at a nominal depth of about 110 m. During the 
following mooring period from 2005 to 2006 this setup did not 
change, but the nominal depths were adjusted to about 50 m and 
60 m, respectively. The first two moorings operated by 
GEOMAR included one 150-kHz and one 75-kHz upward- 
looking ADCP located at about 130 m (200 m) and 620 m 
{700 m) between 2006 to 2008 (2008 to 2009). The upper ADCP 
of the mooring period between 2008 to 2009 was later identified 
by the manufacturer to have electronic problems and could not 
be used after an evaluation of the data records. Unfortunately, 
the very same ADCP was also installed between 2004 to 2005. 
Therefore, data from the shallow ADCP during both mooring 
periods were excluded. In 2009, the design has been changed to 
include one upward-looking 150-kHz ADCP monitoring the full 
EUC depth range and one downward-looking 75-kHz ADCP 
monitoring the intermediate levels. Both ADCPs have been 
installed at a depth of about 210 to 220 m, separated only by a 
few meters along the mooring cable. 
The depth resolution of the ADCPs is defined by vertical bins 
prior to the mooring deployment in the instrument settings. The 
extent of these vertical bins was generally set to 8 m for the 
upward-looking and to 8 or 16 m for the downward-looking 
ADCP. While velocities were not corrected during processing, 
the bin depths were corrected by applying a climatological sound 
speed profile for the instrument’s geographical position 
(Krahmann et al., 2021). Post-processed data were interpolated 
to a vertical resolution of 10 m and current velocities were 40- 
hour low-pass filtered to remove tidal velocities and then 
une 2022 | Valıume 9 | Article 91097