Moored time series of abyssal velocity and temperature from a four-mooring array (M1–M4) deployed across the entrance to the Samoan Passage from July 2012 to February 2014. The array measured the dense bottom flow of Antarctic and North Atlantic origin that passes through the Samoan Passage on its way into the North Pacific. These data are the basis for the analysis of the recent warming and weakening of the abyssal flow reported in Voet et al. (2016).

This repository contains the processed mooring time series, the raw instrument records, and mooring diagrams. It accompanies the publication listed under Citing the data.

---

Description

An array of four moorings (M1–M4) was deployed at the entrance to the Samoan Passage during the first major cruise of the Samoan Passage Abyssal Mixing Experiment (SPAMEX) aboard R/V Revelle in July and August 2012, and recovered during the second SPAMEX cruise aboard R/V Thomas G. Thompson in January and February 2014. The moorings were placed at the same locations as the western four moorings of the six-element array measured in the early 1990s (Rudnick 1997), allowing a direct comparison of the abyssal volume transport across two decades.

Each mooring was instrumented with Aanderaa Recording Current Meters (RCM8) and Sea-Bird SBE39 temperature recorders at depths spanning the abyssal layer below ~3700 m. Two acoustic current meters (Teledyne RDI Doppler Volume Sampler, DVS) and an SBE37 microcat were added to mooring M3, and two ADCPs were attached to the upper part of M3 (the ADCP data return was too poor to be used; the raw ADCP binary files are not archived here, see Notes).

The companion moorings M5–M7, located further east, are not part of this archive.

---

Repository structure

The data are distributed as gzipped tar archives, one per category, so that the internal directory structure is preserved. Top-level files:

.
├── README.md                   This file
├── abstract.md                 Dataset abstract
├── LICENSE                     Creative Commons Zero 1.0 (CC0)
├── proc.tar.gz                 Processed mooring time series (MATLAB .mat)
├── volume_transport.tar.gz     Calculated abyssal volume transport time series
├── raw.tar.gz                  Raw instrument records and field documentation
└── diagrams.tar.gz             Mooring diagrams

Extract an archive with tar xzf <name>.tar.gz, or all at once with for f in *.tar.gz; do tar xzf "$f"; done. Each archive expands to a directory with the following respective structures:

proc/                           Processed mooring time series (MATLAB .mat)
├── m1_original_data.mat         Mooring M1 — all instruments
├── m2_original_data.mat         Mooring M2 — all instruments
├── m3_original_data.mat         Mooring M3 — all instruments
├── m4_original_data.mat         Mooring M4 — all instruments
└── spm1hr.mat                   Hourly array dataset (all moorings) + bathymetry

volume_transport/                Calculated abyssal volume transport time series
├── sp_volume_transport_time_series.mat
└── sp_observed_volume_transport.nc

raw/                             Raw instrument records and field documentation
├── M1/ M2/ M3/ M4/              Per-mooring, per-instrument vendor files
├── SP14_Instrument_time_offsets.txt
└── SPAM_EX_sp12monitor_Aanderaa_recovery_spreadsheet.xlsx

diagrams/                        Mooring diagrams
├── deployed/                    As-deployed schematics (M1–M4, .jpg)
└── recovered/                   As-recovered diagrams (M1–M4, .pdf)

---

Processing code

The code used to process the raw instrument records in raw/ into the calibrated time series in proc/ is available at

https://github.com/gunnarvoet/samoan-passage-rcm-processing

---

Mooring array

Positions, nominal bottom depths, and instrument complement of the four moorings. Instrument depths and record durations are listed in Voet et al. (2016), Table 1.

• Deployment: July 2012 (R/V Revelle, SPAMEX leg 1)
• Recovery: January–February 2014 (R/V Thompson, SPAMEX leg 2)
• Record span: ~22 Jul 2012 – 10 Feb 2014 (≈567 days); most instruments returned more than one year of data.

Instrumentation

• Aanderaa RCM8 — recording current meters, 1-hour vector-averaging sampling. Measure horizontal velocity (speed and direction) and temperature. Nominal temperature accuracy ±3 × 10⁻² °C, resolution ±8 × 10⁻³ °C.
• Sea-Bird SBE39 — temperature recorders, accuracy ±2 × 10⁻³ °C, resolution ±1 × 10⁻⁴ °C.
• Sea-Bird SBE37 MicroCAT — conductivity–temperature recorder (M3 only).
• Teledyne RDI DVS (Doppler Volume Sampler) — single-point acoustic current meters (M3 only); higher-rate records were averaged and downsampled to match the 1-hour RCM8 sampling.

---

Processed data (proc.tar.gz → proc/)

All processed files are MATLAB .mat files. Times are stored as MATLAB datenum (days since 0000-01-00, UTC); yday is the day of year. Velocities are in m s⁻¹, temperatures in °C, depths in m.

m1_original_data.mat … m4_original_data.mat

One file per mooring. Each file contains a single MATLAB structure (named m1, m2, … after the mooring) with mooring metadata and one sub-structure per instrument:

Mooring-level fields

Instrument sub-structures

• s1, s2, … — temperature recorders (SBE39 / SBE37), ordered by depth.
• rcm1, rcm2, … — current meters (RCM8 / DVS), ordered by depth.

The number of instruments varies by mooring (see the table above). Sub-structure fields:

spm1hr.mat

The full array assembled onto a common hourly grid, together with the bathymetry of the mooring section. Contains three variables:

spm — a 1 × 22 structure array, one element per instrument record across all four moorings (current meters and DVS). Fields:

b — bathymetry of the main mooring section: idist, ilon, ilat, itopo are the interpolated distance (km), longitude, latitude and topography (m, negative down) along the section at res = 0.1 km resolution; odist, olon, olat, otopo give the locations of the four moorings.

be — an extended bathymetry section covering a wider distance range (same field names as b, plus dx).

These files load directly in MATLAB. They can also be read in Python with the gvpy package, whose gvpy.io.loadmat returns the structures as nested, dot-addressable dictionaries.

---

Volume transport (volume_transport.tar.gz → volume_transport/)

The abyssal volume transport through the Samoan Passage, computed as the spatial integral of the objectively mapped velocity field between the moorings below 4000 m depth (Voet et al. 2016, their Fig. 3; method following Rudnick 1997). The same time series is provided in two formats.

sp_volume_transport_time_series.mat

A single MATLAB structure with the transport time series in m³ s⁻¹ and time in MATLAB datenum (UTC). 1 Sv = 10⁶ m³ s⁻¹.

sp_observed_volume_transport.nc

The same series as a CF-style netCDF dataset on a single time coordinate (datetime64, UTC) spanning both observational periods. Variables are in m³ s⁻¹ (divide by 10⁶ for Sv); periods without data are filled with NaN.

The time-mean transport over the well-constrained 2012–2014 period is ≈ 5.4 Sv, reduced from ≈ 6.0 Sv in the 1990s (Voet et al. 2016).

---

Raw data (raw.tar.gz → raw/)

Unprocessed instrument records as offloaded in the field, organized by mooring and instrument type. These are the original vendor files preserved for provenance; the scientifically useful, calibrated time series are in proc/.

raw/M<n>/Aanderaa/SN<serial>/   RCM8 current meters: .dsu (binary) and .Asc (ascii)
raw/M<n>/SBE39/SN<serial>/      SBE39 thermistors: .asc, .mat, deploy/recovery .cap logs
raw/M3/SBE37/SN6319/            SBE37 microcat: .asc, .mat, .cap logs
raw/M3/DVS/SN<serial>/          DVS current meters: processed .mat and recovery logs
raw/M3/ADCP/SN<serial>/         ADCP: recovery logs and processed .mat (raw binaries omitted)

Field documentation:

• SP14_Instrument_time_offsets.txt — instrument clock vs. UTC offsets read at recovery (used for clock-drift correction during processing).
• SPAM_EX_sp12monitor_Aanderaa_recovery_spreadsheet.xlsx — Aanderaa RCM recovery log / instrument tracking spreadsheet.

Large raw acoustic binary files (ADCP *.000, DVS *.PD0/*.p20, and the raw ADCP *_raw.mat) are not included here to keep the archive at a reasonable size; the corresponding instruments are documented by their recovery logs and, where used, their processed .mat files.

---

Diagrams (diagrams.tar.gz → diagrams/)

• deployed/ — as-deployed mooring schematics for M1–M4 (.jpg).
• recovered/ — as-recovered mooring diagrams for M1–M4 (.pdf).

---

Notes and data quality

• M1 pressure recorder flooded during deployment; no pressure information exists for this mooring. Given the small standard deviations in pressure (<1 dbar) on the other three moorings, M1 is assumed to have stood upright at the intended instrument depths (Voet et al. 2016).
• M4 RCM8 at 3698 m (SN 9806) returned temperature only; no usable velocity. In spm1hr.mat the velocity fields of this record are filled with NaN.
• M3 ADCPs (upper mooring) returned very poor data because of the low scattering environment at these depths and are not used in the analysis. Their raw binary files are not archived (see raw/M3/ADCP/ recovery logs).
• M3 DVS temperature is not available for the two DVS records (t is NaN); velocity from the DVS was averaged to the 1-hour RCM8 sampling.
• A handful of instruments stopped early due to battery failure; record durations per instrument are given in Voet et al. (2016), Table 1.
• All times are UTC. Clock drifts were corrected during processing using the offsets in raw/SP14_Instrument_time_offsets.txt.

---

Citing the data

This dataset is released into the public domain under CC0 1.0 — you may use it for any purpose without restriction. If you use these data, we request that you cite the accompanying publication:

Voet, G., M. H. Alford, J. B. Girton, G. S. Carter, J. B. Mickett, and J. M. Klymak (2016). Warming and Weakening of the Abyssal Flow through Samoan Passage. Journal of Physical Oceanography, 46(8), 2389–2401. https://doi.org/10.1175/JPO-D-16-0063.1

Related publications on the Samoan Passage abyssal flow:

Alford, M. H., J. B. Girton, G. Voet, G. S. Carter, J. B. Mickett, and J. M. Klymak (2013). Turbulent mixing and hydraulic control of abyssal water in the Samoan Passage. Geophysical Research Letters. https://doi.org/10.1002/grl.50684

Voet, G., J. B. Girton, M. H. Alford, G. S. Carter, J. M. Klymak, and J. B. Mickett (2015). Pathways, Volume Transport, and Mixing of Abyssal Water in the Samoan Passage. Journal of Physical Oceanography. https://doi.org/10.1175/JPO-D-14-0096.1

---

Funding

This work was supported by the U.S. National Science Foundation under grant OCE-1029268.

---

Contact

Gunnar Voet, Scripps Institution of Oceanography, University of California San Diego — gvoet@ucsd.edu

Questions about the data are welcome.