Technical ReportPDF Available

Southern Ocean Time Series (SOTS) Quality Assessment and Control Report - Sediment Trap Particle Fluxes Version 1.0

Authors:
  • Australian Antarctic Partnership Program
Southern Ocean Time Series (SOTS)
Quality Assessment and Control Report -
Sediment Trap Particle Fluxes
Version 1.0
1997-2018
Wynn-Edwards CA, Davies DM, Shadwick EH, Trull TW
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) i
Cathryn A. Wynn-Edwards, Diana M. Davies, Elizabeth H. Shadwick, Thomas W. Trull
Citation
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS
Quality assessment and control report. Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI:
10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
Important disclaimer
CSIRO advises that the information contained in this publication comprises general statements based
on scientific research. The reader is advised and needs to be aware that such information may be
incomplete or unable to be used in any specific situation. No reliance or actions must therefore be
made on that information without seeking prior expert professional, scientific and technical advice. To
the extent permitted by law, CSIRO (including its employees and consultants) excludes all liability to
any person for any consequences, including but not limited to all losses, damages, costs, expenses and
any other compensation, arising directly or indirectly from using this publication (in part or in whole)
and any information or material contained in it.
CSIRO is committed to providing web accessible content wherever possible. If you are having
difficulties with accessing this document please contact csiroenquiries@csiro.au.
ii | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
Foreword
The Southern Ocean Time Series (SOTS) is a Sub-Facility of the Australian Integrated Marine
Observing System (IMOS), funded by the National Collaborative Research Infrastructure Strategy
(NCRIS). It is operated under collaborative arrangements among the CSIRO Oceans and
Atmosphere, Bureau of Meteorology, and University of Tasmania, including via the Antarctic Climate
and Ecosystems Cooperative Research Centre and the Australian Antarctic Program Partnership.
The primary focus is sustained observing of ocean properties and processes important to climate,
carbon cycling, and ocean productivity.
The SOTS Sub-Facility consists of deep ocean moorings deployed in Subantarctic waters southwest
of Tasmania, equipped with autonomous sensors and sample collectors. SOTS moorings are
serviced annually - the existing moorings are recovered and new moorings are deployed. Some
sensor data is transmitted from the moorings via satellite in near real time. Other sensor data and
samples are recovered during the annual service visit.
This report details the quality assessment and control procedures applied to the particle flux data
collected by the Subantarctic Zone (SAZ) sediment trap moorings. The datasets are publicly
available via the AODN Data Portal https://portal.aodn.org.au/search.
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) iii
Contents
FOREWORD ................................................................................................................................................................ II
CONTENTS ................................................................................................................................................................. III
FIGURES ..................................................................................................................................................................... IV
TABLES....................................................................................................................................................................... VI
ACKNOWLEDGMENTS ............................................................................................................................................... VII
EXECUTIVE SUMMARY ............................................................................................................................................. VIII
1 INTRODUCTION .................................................................................................................................................. 1
2 QUALITY ASSURANCE ......................................................................................................................................... 4
3 HIERARCHY OF QC EVALUATION ........................................................................................................................ 5
3.1 CUP INTEGRITY AND REPRESENTATIVENESS .................................................................................................................. 6
3.1.1 Sample cup supernatant pH and salinity values on recovery ........................................................................ 6
ESTIMATI ON OF ANALYTICAL UNCERTA INTIES ....................................................................................................................................7
3.1.2 Sediment trap tilt and current speed ............................................................................................................ 8
3.2 QC PROCEDURES FOR ANALYTICAL RESULTS AND ASSOCIATED FLAG ASSIGNMENTS ............................................................... 9
3.2.1 Sample processing ........................................................................................................................................ 9
3.2.2 Mass flux ..................................................................................................................................................... 10
SAMPLE R ESULT QC TESTS ............................................................................................................................................................10
3.2.3 PIC Coulometry ......................................................................................................................................... 14
ANALYTIC AL QUA LITY C ONTRO L TESTS ............................................................................................................................................14
SAMPLE R ESULT QC TEST S ............................................................................................................................................................16
ESTIMATI ON OF ANALYTICAL UNCERTA INTIES ..................................................................................................................................21
3.2.4 CHN Elemental Analysis ........................................................................................................................... 24
ANALYTIC AL QUA LITY C ONTRO L TEST S .............................................................................................................................................24
SAMPLE R ESULT QC TESTS ............................................................................................................................................................31
ESTIMATI ON OF ANALYTICAL UNCERTA INTIES ..................................................................................................................................39
3.2.5 POC via PC and PIC ................................................................................................................................... 42
SAMPLE R ESULT QC TESTS ............................................................................................................................................................42
ESTIMATI ON OF ANALYTICAL UNCERTA INTIES ..................................................................................................................................46
3.2.6 Biogenic silica alkaline digest ................................................................................................................... 47
ANALYTIC AL QUA LITY C ONTRO L TESTS ............................................................................................................................................47
SAMPLE R ESULT QC TEST S ............................................................................................................................................................49
ESTIMATI ON OF ANALYTICAL UNCERTA INTIES ..................................................................................................................................53
3.2.7 Total mass balance test .............................................................................................................................. 54
3.3 FINAL ASSESSMENT OF RESULTS ............................................................................................................................... 54
3.4 MULTI-YEAR REVIEW OF RESULTS AND ASSOCIATED FLAGS ............................................................................................ 55
REFERENCES ..............................................................................................................................................................58
iv | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
Figures
Figure 1. Mass flux QC test sequence and results. ....................................................................... 10
Figure 2. %CaCO3 of reference material CaCO3 over time as chemical analysis control chart.
Solid line = mean, dash-dotted line = 2 SD. .................................................................................. 15
Figure 3: %CaCO3 of reference material 1999_54_1500 over time as chemical analysis control
chart. Solid line = mean, dash-dotted line = 2 SD. ........................................................................ 15
Figure 4: %CaCO3 of reference material PACS-2 over time as chemical analysis control chart.
Solid line = mean, dash-dotted line = 2 SD. .................................................................................. 15
Figure 5: %CaCO3 of duplicates over time as chemical analysis control chart. Dash-dotted line =
5%. ................................................................................................................................................. 16
Figure 6: %PIC flux QC test sequence and results......................................................................... 17
Figure 7: %C (top) and %N (bottom) of reference material PACS-2 over time as chemical analysis
control chart. Solid line = mean, dash-dotted line = 2 SD. ........................................................... 28
Figure 8: %C (top) and %N (bottom) of reference material Acetanilide over time as chemical
analysis control chart. Solid line = mean, dash-dotted line = 2 SD. .............................................. 29
Figure 9: %C (top) and %N (bottom) of duplicates over time as chemical analysis control chart.
Dash-dotted line = 5%. .................................................................................................................. 30
Figure 10:%PC flux QC test sequence and results. ....................................................................... 31
Figure 11: %PN flux QC test sequence and results. ...................................................................... 32
Figure 12: %POC flux QC test sequence and results. .................................................................... 42
Figure 13: %BSi of reference material 1999_54_1500 over time as chemical analysis control
chart. Solid line = mean, dash-dotted line = 2 SD. ........................................................................ 47
Figure 14: %BSi of reference material PACS-2 over time as chemical analysis control chart. Solid
line = mean, dash-dotted line = 2 SD. ........................................................................................... 48
Figure 15: %BSi of duplicates over time as chemical analysis control chart. Dash-dotted line =
5%. ................................................................................................................................................. 48
Figure 16: %BSi flux QC test sequence and results. ...................................................................... 49
Figure 17. Mass flux data with associated QC flags. A. 1000m (filled circles), 1050m upward
pointing triangles and 500m (open circles). B. 2000m. C. 3800m (downward pointing triangles)
and 2850m (crosses). QC flags are indicated by marker colour; QC flag 1 = blue, QC flag 2 =
green and QC flag 3 = orange........................................................................................................ 55
Figure 18. POC percentage flux data with associated QC flags. A. 1000m (filled circles), 1050m
upward pointing triangles and 500m (open circles). B. 2000m. C. 3800m (downward pointing
triangles) and 2850m (crosses). QC flags are indicated by marker colour; QC flag 1 = blue, QC
flag 2 = green and QC flag 3 = orange. .......................................................................................... 55
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) v
Figure 19. PIC percentage flux data with associated QC flags. A. 1000m (filled circles), 1050m
upward pointing triangles and 500m (open circles). B. 2000m. C. 3800m (downward pointing
triangles) and 2850m (crosses). QC flags are indicated by marker colour; QC flag 1 = blue, QC
flag 2 = green and QC flag 3 = orange. .......................................................................................... 56
Figure 20. PC percentage flux data with associated QC flags. A. 1000m (filled circles), 1050m
upward pointing triangles and 500m (open circles). B. 2000m. C. 3800m (downward pointing
triangles) and 2850m (crosses). QC flags are indicated by marker colour; QC flag 1 = blue, QC
flag 2 = green and QC flag 3 = orange. .......................................................................................... 56
Figure 21. PN percentage flux data with associated QC flags. A. 1000m (filled circles), 1050m
upward pointing triangles and 500m (open circles). B. 2000m. C. 3800m (downward pointing
triangles) and 2850m (crosses). QC flags are indicated by marker colour; QC flag 1 = blue, QC
flag 2 = green and QC flag 3 = orange. .......................................................................................... 57
Figure 22. BSi percentage flux data with associated QC flags. A. 1000m (filled circles), 1050m
upward pointing triangles and 500m (open circles). B. 2000m. C. 3800m (downward pointing
triangles) and 2850m (crosses). QC flags are indicated by marker colour; QC flag 1 = blue, QC
flag 2 = green and QC flag 3 = orange. .......................................................................................... 57
vi | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
Tables
Table 1: Instrument deployment details......................................................................................... 2
Table 2. Mass flux QC thresholds - 1000m ................................................................................... 11
Table 3. Mass flux QC thresholds - 2000m ................................................................................... 12
Table 4. Mass flux QC thresholds - 3800m ................................................................................... 13
Table 5. %PIC flux QC thresholds - 1000m .................................................................................... 18
Table 6.%PIC flux QC thresholds - 2000m ..................................................................................... 19
Table 7. %PIC flux QC thresholds - 3800m .................................................................................... 20
Table 8. POC/PN flux QC thresholds - 1000m ............................................................................... 25
Table 9. POC/PN flux QC thresholds - 2000m ............................................................................... 26
Table 10. POC/PN flux QC thresholds - 3800m ............................................................................. 27
Table 11. %PC flux QC thresholds - 1000m ................................................................................... 33
Table 12. %PN flux QC thresholds - 1000m .................................................................................. 34
Table 13. %PC flux QC thresholds - 2000m ................................................................................... 35
Table 14. %PN flux QC thresholds - 2000m .................................................................................. 36
Table 15. %PC flux QC thresholds - 3800m ................................................................................... 37
Table 16. %PN flux QC thresholds - 3800m .................................................................................. 38
Table 17. %POC flux QC thresholds - 1000m ................................................................................ 43
Table 18. %POC flux QC thresholds - 2000m ................................................................................ 44
Table 19. %POC flux QC thresholds - 3800m ................................................................................ 45
Table 20. %BSi flux QC thresholds - 1000m .................................................................................. 50
Table 21. %BSi flux QC thresholds - 2000m .................................................................................. 51
Table 22. %BSi flux QC thresholds - 3800m .................................................................................. 52
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) vii
Acknowledgments
We acknowledge support from the following agencies: the Australian Antarctic Program Partnership
(AAPP), the Antarctic Climate and Ecosystems Cooperative Research Centre (ACE CRC), the
Integrated Marine Observing System (www.imos.org.au), University of Tasmania (UTAS), Bureau of
Meteorology (BoM), the Marine National Facility (MNF) and the Australian Antarctic Division (AAD).
We also acknowledge the support of the CSIRO Moored Sensor Systems team.
SOTS is a member of the OceanSITES global network of time series observatories
(www.OceanSITES.org).
viii | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
Executive summary
The Southern Ocean Time Series (SOTS) Observatory located near 140E and 47S provides high
temporal resolution observations in subantarctic waters. It is focused on the subantarctic Zone
because waters formed at the surface in this region slide under warmer subtropical and tropical
waters, carrying CO2 and heat into the deep ocean, where it is out of contact with the atmosphere.
This process also supplies oxygen for deep ocean ecosystems, and exports nutrients that fuel ~70%
of global ocean primary production. This region is also the boundary between the nutrient rich
waters of the Southern Ocean and the oligotrophic subtropical gyres to the north. These processes
are sensitive to climate change, but the probable nature and impacts are not yet known.
This report details the quality assessment and control procedures applied to the particle flux data
collected by the Subantarctic Zone (SAZ) moorings. The datasets are publicly available via the IMOS
Data Portal AODN https://portal.aodn.org.au/search. This report should be consulted when using
the data.
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 1
1 Introduction
Detailed descriptions of mooring designs, locations and sample collections can be found in the SOTS
annual reports. A summary of instrument models, serial numbers, depth and deployments is listed
in table 1. The SOTS annual reports (available at https://catalogue-
imos.aodn.org.au/geonetwork/srv/eng/metadata.show?uuid=afc166ce-6b34-44d9-b64c-
8bb10fd43a07) are divided into three parts:
Report 1. Overview, details mooring voyages, dates, locations, designs and instruments
Report 2. Samples, details the sample collections, and
Report 3. Sensors, contains descriptions of sensors on the moorings, and the QC procedures applied
to data from those sensors.
This report outlines the tests performed on sediment trap particle flux data that led to the allocation
of quality control flags, and uncertainty estimates.
Data flags in the on-line netcdf files (available from AODN.org.au) follow the Argo table 2A Flag
convention (equivalent to IMOS Standard Flag, OCEANSITES and IODE flag convention) and are as
follows:
0 No QC was performed, 1 good data, 2 probably good, 3 bad data that are potentially correctable,
4 bad data, 5 value changed, 6 not used, 7 not used, 8 interpolated value, 9 missing value.
Final QC flags are a combination of sample QC flags, mass balance test results, and component flags.
The final QC flag is always the highest flag from all tests.
The quality of the measurements derives from three main aspects:
i. Quality assurance of sample collection i.e. the preparation and deployment of the
sediment traps. This is described in more detail in the annual Sample Report. It does
not lead to a formal uncertainty estimate or a flag but is important to understand the
overall fidelity of the observations.
ii. Quality assessment of the recovered samples i.e. whether the cups were recovered as
expected; may have had excessive organic material in it, etc. This is described in section
3.1 and further elaborated on in the sections on individual chemical analyses. As with
the quality assurance this does not lead to a formal uncertainty estimate, but the result
of this assessment will be considered in the assignment of QC flags. This section also
describes the uncertainties that arise from sample splitting and factors with potential to
affect sample quality but with unknown correction factors, i.e. trap tilt and ocean
current. Tilt and current data on their own do not lead to assignment of flags but are
important to understand the overall fidelity of the observations.
iii. Analytical error estimates for each analyte are estimated by combining information from
analytical calibrations, standard reference materials, replicate analyses, and blank
determinations. Relative standard uncertainty of accuracy and precision are combined
and multiplied with a coverage factor 2 (expanded uncertainty) to represent the 95%
confidence interval.
2 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
We recommend the use of data with associated QC flags 1 or 2 and suggest contacting the facility
for futher information before using data with QC flags 3.
The procedures described in this document have been applied to the deployments listed in Table 1.
Most of these occurred at the core SOTS site at 47°S, however additional deployments at 45 and
46°S are also addressed. The climatological values for all sites are drawn from the core SOTS site
between 1997 and 2017 and thus the applicability is less certain at the other locations. Most traps
were deployed at three depths, nominally 1000, 2000 and 3800m (i.e. 700m above the sea floor).
Some additional trap deployments occurred at 500m and 2850m. For these depths the
climatological tests are based on the 1000m and 3800m thresholds, respectively.
Table 1: Instrument deployment details
Deployments and depth
Producer
Model
Serial No.
Long
SAZ47-01-1997 (1000m)
PARFLUX
Mark78H-21
10583
142.07
SAZ47-01-1997 (2000m)
PARFLUX
Mark78H-21
801
142.07
SAZ47-01-1997 (3800m)
PARFLUX
Mark78H-21
1021
142.07
SAZ47-02-1998 (1000m)
PARFLUX
Mark78H-21
10583
142.09
SAZ47-02-1998 (2000m)
PARFLUX
Mark78H-21
801
142.09
SAZ47-02-1998 (3800m)
PARFLUX
Mark78H-21
1021
142.09
SAZ47-03-1999 (1000m)
PARFLUX
Mark78H-21
10583
142.10
SAZ47-03-1999 (2000m)
PARFLUX
Mark78H-21
2241
142.10
SAZ47-03-1999 (3800m)
PARFLUX
Mark78H-21
1513
142.10
SAZ47-04-2000 (1000m)
PARFLUX
Mark78H-21
10583
142.04
SAZ47-04-2000 (2000m)
PARFLUX
Mark78H-21
2241
142.04
SAZ47-07-2003 (500m)
PARFLUX
Mark78H-21
11640-01
141.65
SAZ47-07-2003 (1000m)
PARFLUX
Mark78H-21
11640-02
141.65
SAZ47-07-2003 (2000m)
PARFLUX
Mark78H-21
2241
141.65
SAZ47-08-2004 (500m)
PARFLUX
Mark78H-21
11640-02
141.45
SAZ47-08-2004 (1000m)
PARFLUX
Mark78H-13
11742-01
141.45
SAZ47-08-2004 (2000m)
PARFLUX
Mark78H-13
11741-01
141.45
SAZ47-09-2005 (1000m)
Prime Focus
Indented Rotating Sphere Trap
IRS-2
141.84
SAZ47-09-2005 (1000m)
PARFLUX
Mark78H-21
10583
141.84
SAZ47-09-2005 (2000m)
PARFLUX
Mark78H-21
11649-01
141.84
SAZ47-11-2008 (1000m)
Prime Focus
Indented Rotating Sphere Trap
IRS-1
145.64
SAZ47-11-2008 (1000m)
PARFLUX
Mark78H-21
11741-01
145.64
SAZ47-11-2008 (2000m)
PARFLUX
Mark78H-21
11640-01
145.64
SAZ47-11-2008 (2900m)
PARFLUX
Mark78H-21
11649-01
145.64
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 3
Deployments and depth
Producer
Model
Serial No.
Long
SAZ47-11-2008 (2900m)
Prime Focus
Indented Rotating Sphere Trap
IRS-2
145.64
SAZ47-12-2009 (1000m)
Prime Focus
Indented Rotating Sphere Trap
IRS-1
141.66
SAZ47-12-2009 (3800m)
PARFLUX
Mark78H-21
12419-03
141.66
SAZ47-13-2010 (1000m)
PARFLUX
Mark78H-21
11741-01
141.65
SAZ47-13-2010 (2000m)
PARFLUX
Mark78H-21
11649-01
141.65
SAZ47-13-2010 (3900m)
PARFLUX
Mark78H-21
11640-01
141.65
SAZ47-14-2011 (1000m)
Prime Focus
Indented Rotating Sphere Trap
IRS-1
141.82
SAZ47-14-2011 (3900m)
PARFLUX
Mark78H-21
12419-03
141.82
SAZ47-15-2012 (1000m)
PARFLUX
Mark78H-21
11741-01
141.68
SAZ47-15-2012 (1050m)
Prime Focus
Indented Rotating Sphere Trap
IRS-2
141.68
SAZ47-15-2012 (2000m)
PARFLUX
Mark78H-21
11649-01
141.68
SAZ47-15-2012 (3900m)
PARFLUX
Mark78H-21
11640-01
141.68
SAZ47-16-2013 (1050m)
Prime Focus
Indented Rotating Sphere Trap
IRS-2
141.82
SAZ47-16-2013 (3900m)
PARFLUX
Mark78H-21
12419-03
141.82
SAZ47-17-2015 (1000m)
PARFLUX
Mark78H-21
12933-01
141.66
SAZ47-17-2015 (1050m)
PARFLUX
Mark78H-21
11649-01
141.66
SAZ47-17-2015 (2000m)
PARFLUX
Mark78H-21
11741-01
141.66
SAZ47-17-2015 (3800m)
PARFLUX
Mark78H-21
11640-01
141.66
SAZ47-18-2016 (3800m)
PARFLUX
Mark78H-21
12419-03
141.84
SAZ47-19-2017 (1000m)
PARFLUX
Mark78H-21
11741-01
142.31
SAZ47-19-2017 (2000m)
PARFLUX
Mark78H-21
11640-01
142.31
SAZ47-19-2017 (3800m)
PARFLUX
Mark78H-21
11649-01
142.31
SAZ47-20-2018 (1000m)
PARFLUX
Mark78H-21
12419-01
141.79
SAZ47-20-2018 (2000m)
PARFLUX
Mark78H-21
12419-02
141.79
SAZ47-20-2018 (3800m)
PARFLUX
Mark78H-21
12993-01
141.79
4 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
2 Quality assurance
Prior to deployment the sediment traps are completely disassembled (including separating the
carousel rotator and stator plate; and removing the baffle) and cleaned. The motor shaft quad rings
are checked and replaced if necessary. New batteries are installed after all tests are complete, just
prior to loading onto the ship.
Sediment trap cups are filled with a collection brine made up of 0.8µm GF/F filtered seawater,
collected close to the SOTS site. NaCl is added to increase density (5 g L-1) and improve retention in
the cups, as well as Na2B4O7 10 H2O (2 g L-1) to buffer pH and reduce carbonate dissolution, SrCl2 6
H2O (0.22 g L-1) for preservation of acantharians, and HgCl2 (3 g L-1) as biocide to prevent particle
remineralization. Brine salinity in the cup prior to deployment is around 40 psu.
The SAZ sediment trap mooring is deployed anchor last (so that the Parflux traps are towed with the
carousel towards the ship during deployment) and normally recovered starting from the top of the
mooring (so that the traps are towed with the opening of the funnel towards the ship during
recovery). Starting in 2020 all sediment traps are fitted with a protective ring around the cups to
reduce the risk of cups breaking off during deployment and recovery. After recovery of the mooring
the sediment traps are hosed down with fresh water and allowed to settle for at least 24 hours.
Each cup has ~ 20% of supernatant removed through the filling hole to reduce the risk of spillage of
hazardous brine during removal of cups from the trap carousel. The sample cups are then capped
and refrigerated in the dark until transfer to the onshore laboratory.
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 5
3 Hierarchy of QC Evaluation
QC test thresholds are based on averages and standard deviations of data collected between 1997
and 2017 at 47°S by McLane Parflux sediment traps. The same QC thresholds are applied to data
collected at 45 and 46°S and to data collected by University of Washington / Prime Focus sediment
traps Indented Rotating Sphere (IRS) traps.
Mass flux and each component flux are
subjected to seasonality tests and depth
comparisons. A mass balance test is
performed per cup. If a component fails a
test and no analytical, sample processing or
collection condition could explain the result
then the final QC flag will be 2. If multiple
components of a sample cup fail tests and
sample condition information, e.g. tilt and
current, supernatant pH and salinity,
indicate possible compromise, then all final
QC flags for the affected cup may be set to
3.
Component QC flags are based on analytical
quality and QC tests, unless overridden by
decisions made earlier during sample
processing or later during final QC
assessment.
Cup supernatant pH and salinity post
recovery may indicate potential washout.
Sediment trap tilt and ocean current speed
may influcence collection efficiency. This
information on its own does not lead to QC
flags but may be used in conjuction with
other cup results to inform sample and
component QC flags.
The integrity of each cup (i.e. exposed
to funnel, intact on recovery, no
compromising material inside, etc.)
leads to a sample QC flag.
Cup integrity
Supernatant
pH and
salinity
Trap tilt and
current
speed
Sample
processing
CHN QC flag
POC QC flag
PIC QC flag BSi QC flag Mass QC flag
Decisions made during sample
processing will lead to QC flags
that carry through to all
components and override QC
test result flags. Examples
include: not enough material (QC
flag 9 for components), or little
material increasing uncertainty
(QC flag 2).
Mass flux and component flux QC flag
6 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
3.1 Cup integrity and representativeness
Every cup is assigned a sample QC flag which signifies whether the sediment trap functioned
correctly, exposing the cup to the collection funnel, and whether the cup was recovered intact and
sealed. If the sediment trap functioned correctly the sample QC flag is 1. Collection cups that were
not exposed to the funnel due to controller malfunction, premature low battery shut down and/or
early recovery of the mooring or cups lost during deployment or on recovery are flagged 9, missing
data. Cups that were recovered physically damaged or leaking are flagged 4, bad data. Cups
recovered under the funnel are assigned sample QC flag 2 (because fluxes have been observed to
fall within normal ranges and departure from this condition are highlighted by subsequent QC tests).
Cups with unusual conditions, e.g. disintegrated organism, are noted in the annual sample reports
and maybe assigned a minimum flag of 2 on this basis.
3.1.1 Sample cup supernatant pH and salinity values on recovery
Supernatant pH and salinity are a possible indicator of brine washout but not necessarily washout
of material. Both are measured before deployment and after recovery. The two tracers have
different properties. Salinity should be conservative with respect to the sample, whereas the pH
may be affected by the sample. Because it is unclear when the washout may have occurred and
whether particles were also washed out, these results are not currently used to assign quality flags
to the particle fluxes. The data is reported for each deployment in the annual SOTS Report 2.
Samples, and we recommend that this information be consulted and considered in use of the sample
flux data. pH and salinity data are assigned QC flags, which reflect the quality of the measurements
only.
About 10% of all collected samples are usually run in duplicates. Where available, duplicate sample
results are recorded in a control chart. The percent difference between duplicates from their mean
is plotted over time.
If either more than one reference material result fell outside the control limits, or if the relative
percent difference of duplicates was more than 5% (Conte et al. 2001, Martiny et al. 2014), the
affected duplicates and a subset of samples of the affected run are repeated. If the repeated
duplicates or the repeated single sample results deviate more than 5% from previous results, all
previous samples are re-run. Otherwise results of the new set of duplicates and the average of the
repeated and previous single results are accepted and the data is flagged as 1.
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 7
ESTIMATION OF ANALYTICAL UNCERTAINTIES
Analytical measurement uncertainty is calculated by combining accuracy and precision estimates.
The underlying equations are given below along with the long-term average expanded
measurement uncertainty.
Accuracy is assessed with the help of reference material, pH buffer (precision ± 0.01) and salinity
reference solution (precision ± 1%), and calculated as follows:
(1)  

 
Precision is assessed with the help of sample duplicates, and calculated as follows:
(2)  
 
The combined measurement uncertainty is calculated as follows:
(3) 
The combined measurement uncertainty is reported as expanded uncertainty with a coverage
factor of 2, to represent a 95% level of confidence.
(4)   
The calculations for salinity measurements follow the same principle.
The long-term average expanded uncertainty for pH measurements is 0.03 or 2.89%.
The long-term average expanded uncertainty for salinity measurements is 0.02 or 2.01%.
This estimate is relative to the uncertainty of the reference buffer of ± 0.01 pH units and ± 1% for
salinity and these uncertainties dominate the overall uncertainty.
The overall pH uncertainty =   =
0.03 or 2.9%
(pH buffers 7.01 and 9.18)
The overall salinity uncertainty =  = 0.022 or
2.2%
(Ellison and Williams 2012, Rees et al. 2018)
8 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
3.1.2 Sediment trap tilt and current speed
While tilt and ocean current may affect collection efficiency, this data is not currently used for
sample results QC because of the uncertain relationships between these variables and sample trap
collection efficiencies. There is no clear consensus regarding the degree to which sediment trap tilt
negatively / positively affects trapping efficiency. Over- and under-trapping have been reported for
cylindrical traps tilting more than (Gardner 1985), and conical traps tilting 5 20° (Chiswell and
Nodder 2015), respectively.
The moorings carry current meters at around 1200m depth. Tilt meters are included in the
controllers of the 1000m and 2000m trap. Current speeds of less than 10 cm s-1 are considered
negligible for trapping efficiency (Baker et al. 1988, Honjo et al. 1992).
Sediment trap tilt and current data is reported in the annual SOTS Report 3. Sensors, and we
recommend that this information be consulted and considered when using the sample flux data.
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 9
3.2 QC procedures for analytical results and associated flag assignments
3.2.1 Sample processing
Observations made during sample processing may inform mass flux and component flux flags
irrespective of analytical measurement quality. Adverse sample conditions are noted before sample
processing and can include excessive organic material in the cup, such as a large fish or fish scales
or a bad smell indicating detrimental bacterial activity. Problems during sample processing can
include loss of sample during filtration.
Samples masses of less than 20mg dry material are imprecise because of the difficulty of recovering
all material from the cup and losses due to decanting. The masses of such samples are generally
flagged as 2 - probably good and this flag is also applied to component fluxes in mg or mol m-2 yr-1.
For sample masses of less than 10mg, no further analysis is carried out and the material is archived
on the filter and only the mass flux is reported (with flag 2). For these small samples, the values of
unanalyzed components are reported as NaN and flagged 9 no data.
Sample splitting
The <1mm fraction is re-suspended and split 10 ways using a McLane wet sample divider WSD-10
(www.Mclanelabs.com) where there was adequate sediment, i.e. at least 10mg. Three splits are
archived and the remaining seven re-combined, filtered onto a Sterlitech 0.4μm 47mm filter and
dried to constant weight at 60°C. The dried solids are then weighed to provide mass flux data and
are available for analysis (specifically the total filter + dried sediment is weighed and the pre-
weighed filter mass is subtracted). The dried sediment is scraped off the filter and ground with
mortar and pestle. This splitting process has a precision of ~8% (Trull et al. 2001). Since 7 of these
10 splits are combined prior to chemical component analysis, the error associated with splitting for
the combined <1mm fraction is 8% / √7 = 3%. This uncertainty is not included in the final 95%
confidence interval for analytical results, because it affects all analytical results similarly.
10 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
3.2.2 Mass flux
The mass flux results are initially given the same flag as the sample cup, which depends on the
performance of the sediment trap during sample collection and will be revised if sample processing
or adverse sample collection conditions further affected sample integrity. Cups that have failed QC
tests are subjected to closer scrutiny. For this purpose, processing notes and sample photos are
consulted. Special conditions of each sample, such as large quantities of soft bodied >1mm
organisms that may break up during sieving can contribute disproportionately to the mass.
SAMPLE RESULT QC TESTS
Mass flux results are subjected to the following QC tests:
i. Cup mass flux seasonality test
ii. Cup-to-cup mass flux QC comparison across depth
Each cup mass flux result in mg m-2 d-1 is compared to the time-series average appropriate to the
time of year of the cup collection interval and depth. A value outside of 3 standard deviations from
the time-series mean is initially flagged 2. If the same cup number at the depth below (or above in
the case of 3800m) failed its seasonality test, then the flag for both cups remains at 2. If the cup at
the next depth passes its seasonality test, then the flag of the cup that failed the seasonality test is
advanced to 3 (Fig 1). This allows for unexpected results to remain flagged 2, probably good, if the
pattern is seen at multiple depths and therefore more likely to be real.
Mass flux QC flags 2 and 3 are applied to all results based on mass flux calculations (unless higher
flag occurs subsequently from chemical analyses). Component percentage flux results are assumed
unaffected, i.e. composition is considered independent of collection efficiency.
Cup seasonality mass flux
Depth comparison test
Test fail
Test pass
Cup that failed the seasonality test,
and who’s counterpart at the next
depth passed its seasonality test is
flagged 3.
Cup that failed the seasonality test, and who’s
counterpart at the next depth also failed its
seasonality test is flagged 2.
QC flag 1
Figure 1. Mass flux QC test sequence and results.
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 11
Test conditions:
Due to large standard deviations and the resulting negative numbers as lower limit for 1000m
results, we chose a seasonally generalized lower limit of 0.
Table 2. Mass flux QC thresholds - 1000m
1000m (1997 2017, 47°S)
Time of integration
Average ± 1 std
[mg m-2 d-1]
Lower limit (average
3 std)
Upper limit (average
+ 3 std)
21st June 4th July
27.4 ± 15.6
0
74.2
5th July 18th July
25.5 ± 12.1
0
61.8
19th July 1st Aug
25.1 ± 10.5
0
56.7
2nd Aug 15th Aug
22.6 ± 8.7
0
48.8
16th Aug 29th Aug
20.5 ± 11.8
0
55.9
30th Aug 12th Sept
22.4 ± 12.8
0
60.6
13th Sept 26th Sept
29.1 ± 17.0
0
80.0
27th Sept 10th Oct
40.6 ± 32.2
0
137.2
11th Oct 24th Oct
61.7 ± 46.5
0
201.0
25th Oct 7th Nov
75.5 ± 47.2
0
217.2
8th Nov 21st Nov
65.0 ± 38.7
0
181.1
22nd Nov 5th Dec
48.6 ± 37.9
0
162.2
6th Dec 19th Dec
43.0 ± 38.3
0
157.9
20th Dec 2nd Jan
45.5 ± 41.6
0
170.4
3rd Jan 16th Jan
40.7 ± 31.9
0
136.3
17th Jan 30th Jan
47.4 ± 39.0
0
164.5
31st Jan 13th Feb
53.1 ± 56.7
0
223.1
14th Feb 27th Feb
41.2 ± 45.5
0
177.8
28th Feb 13th Mar
39.0 ± 31.6
0
133.9
14th Mar 27th Mar
42.1 ± 29.7
0
131.1
28th Mar 10th Apr
36.8 ± 26.7
0
116.9
11th Apr 24th Apr
29.1 ± 22.9
0
97.9
25th Apr 8th May
27.8 ± 21.2
0
91.3
9th May 22nd May
29.0 ± 18.1
0
83.2
23rd May 5th June
30.4 ± 14.5
0
73.8
6th June 20th June
30.5 ± 13.1
0
69.7
12 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
Negative or near 0 (<1) lower limits were replaced by a lower limit of 0.
Table 3. Mass flux QC thresholds - 2000m
2000m (1997 2017, 47°S)
Time of integration
Average ± 1 std
[mg m-2 d-1]
Lower limit (average
3 std)
Upper limit (average
+ 3 std)
21st June 4th July
39.1 ± 13.4
0
79.3
5th July 18th July
33.8 ± 10.4
2.73
64.9
19th July 1st Aug
31.5 ± 12.4
0
68.6
2nd Aug 15th Aug
32.0 ± 8.8
5.59
58.5
16th Aug 29th Aug
31.2 ± 7.5
8.74
53.6
30th Aug 12th Sept
31.6 ± 10.5
0
63.2
13th Sept 26th Sept
32.0 ± 15.6
0
78.9
27th Sept 10th Oct
35.5 ± 21.1
0
98.8
11th Oct 24th Oct
47.0 ± 28.9
0
133.7
25th Oct 7th Nov
66.7 ± 34.4
0
170.0
8th Nov 21st Nov
90.2 ± 40.9
0
212.9
22nd Nov 5th Dec
107.0 ± 60.6
0
288.8
6th Dec 19th Dec
102.1 ± 68.5
0
307.5
20th Dec 2nd Jan
92.2 ± 57.2
0
263.7
3rd Jan 16th Jan
90.9 ± 47.6
0
233.6
17th Jan 30th Jan
76.5 ± 48.2
0
221.1
31st Jan 13th Feb
73.0 ± 63.7
0
264.2
14th Feb 27th Feb
64.9 ± 49.8
0
214.4
28th Feb 13th Mar
52.7 ± 31.1
0
146.0
14th Mar 27th Mar
45.4 ± 24.4
0
118.7
28th Mar 10th Apr
42.9 ± 24.2
0
115.5
11th Apr 24th Apr
39.7 ± 20.1
0
100.1
25th Apr 8th May
40.7 ± 18.0
0
94.7
9th May 22nd May
41.7 ± 17.4
0
93.9
23rd May 5th June
43.9 ± 19.5
0
102.4
6th June 20th June
44.2 ± 22.0
0
110.3
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 13
Negative or near 0 (<1) lower limits were replaced by a lower limit of 0.
Table 4. Mass flux QC thresholds - 3800m
3800m (1997 2017, 47°S)
Time of integration
Average ± 1 std
[mg m-2 d-1]
Lower limit (average
3 std)
Upper limit (average
+ 3 std)
21st June 4th July
39.3 ± 16.1
0
87.4
5th July 18th July
38.6 ± 13.1
0
78.0
19th July 1st Aug
37.3 ± 13.9
0
79.0
2nd Aug 15th Aug
33.6 ± 8.9
7.1
60.2
16th Aug 29th Aug
31.1 ± 6.7
10.9
51.3
30th Aug 12th Sept
29.5 ± 8.0
5.4
53.5
13th Sept 26th Sept
31.7 ± 10.4
0
62.9
27th Sept 10th Oct
35.4 ± 13.3
0
75.3
11th Oct 24th Oct
40.9 ± 13.6
0
81.7
25th Oct 7th Nov
51.5 ± 9.7
22.4
80.6
8th Nov 21st Nov
64.3 ± 13.7
23.2
105.4
22nd Nov 5th Dec
68.3 ± 22.8
0
136.8
6th Dec 19th Dec
70.7 ± 29.5
0
159.3
20th Dec 2nd Jan
77.4 ± 27.4
0
159.6
3rd Jan 16th Jan
85.7 ± 28.8
0
172.0
17th Jan 30th Jan
102.3 ± 43.7
0
233.4
31st Jan 13th Feb
108.0 ± 50.7
0
260.0
14th Feb 27th Feb
94.4 ± 44.9
0
229.0
28th Feb 13th Mar
72.9 ± 21.5
8.2
137.5
14th Mar 27th Mar
59.9 ± 13.3
20.0
99.7
28th Mar 10th Apr
55.9 ± 11.3
21.9
89.9
11th Apr 24th Apr
51.9 ± 13.0
12.9
90.9
25th Apr 8th May
52.5 ± 13.1
13.2
91.8
9th May 22nd May
54.9 ± 13.4
14.5
95.2
23rd May 5th June
55.1 ± 18.2
0
109.8
6th June 20th June
58.2 ± 20.3
0
119.3
14 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
3.2.3 PIC Coulometry
QC flags for component fluxes are based on the quality of the chemical analyses and then subjected
to QC tests similar to those outlined for mass fluxes. The quality of the chemical analysis will only
overrule the QC flag based on QC tests if the quality of the chemical analysis was compromised and
the sample could not be re-run.
ANALYTICAL QUALITY CONTROL TESTS
Analytical runs contain at least one of the following three reference materials, PACS-2 (sediment),
which is low in PIC, 1999_54_1500m, a bulk trap sample quality control material collected earlier in
the study and CaCO3 as standard for the calibration curve as well as reference material throughout
the analysis run. %PIC results of these reference materials are recorded in a quality control chart.
Upper and lower quality control limits for reference material %PIC (w/w) results were set to 2 times
standard deviation.
About 10% of all collected samples are usually run in duplicates. Where available, duplicate sample
results are recorded in a control chart. The percent difference between duplicates from their mean
is plotted over time.
If either i) more than one reference material result fell outside the control limits, or ii) if the relative
percent difference of duplicates was more than 5% (Conte et al. 2001, Martiny et al. 2014), the
affected duplicates and a subset of samples of the affected run are repeated. If the repeated
duplicates or the repeated single sample results deviate more than 5% from previous results, all
previous samples are re-run. Otherwise results of the new set of duplicates and the average of the
repeated and previous single results are accepted and the data is flagged as 1.
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 15
Figure 2. %CaCO3 of reference material CaCO3 over time as chemical analysis control chart. Solid line = mean, dash-
dotted line = 2 SD.
Figure 3: %CaCO3 of reference material 1999_54_1500 over time as chemical analysis control chart. Solid line = mean,
dash-dotted line = 2 SD.
Figure 4: %CaCO3 of reference material PACS-2 over time as chemical analysis control chart. Solid line = mean, dash-
dotted line = 2 SD.
96
98
100
102
104
106
108
010 20 30 40 50 60
%CaCO3
Running number
CaCO3reference material
0
5
10
15
20
25
010 20 30 40 50 60 70 80 90 100
%CaCO3
Running number
1999_54_1500 reference material
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0 5 10 15 20 25 30 35 40 45
%CaCO3
Running number
PACS-2 reference material
16 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
Figure 5: %CaCO3 of duplicates over time as chemical analysis control chart. Dash-dotted line = 5%.
SAMPLE RESULT QC TESTS
After quality of analytical results has been assured, particle composition values are subjected to
quality control tests similar to those described for mass fluxes based on time series averages and
3 standard deviations as upper and lower limit (Fig 6).
Cups that have failed QC tests are subjected to closer scrutiny. For this purpose, processing notes
and sample photos are consulted. Photos showing a cover of white sand on the filter will explain an
unusually high PIC result. Very high concentrations of calcifying organisms, often visible as white
sand on the <1mm fraction filters, will lead to high %PIC and thus high %C results. These are
situations that can explain unexpected results that will lead to QC test failures but will not lead to a
high QC flag.
If cups fail all QC tests and processing notes have indicated potentially compromising conditions,
then the final QC flag will be 4.
i. Cup %PIC seasonality test
ii. Cup-to-cup % PIC QC comparison across depth
-6.0%
-4.0%
-2.0%
0.0%
2.0%
4.0%
6.0%
010 20 30 40 50 60 70 80 90 100
%CaCO3
Running number
%CaCO3duplicates
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 17
Cup seasonality %PIC test
Depth comparison test
Test fail
Test pass
Cup that failed the seasonality test,
and who’s counterpart at the next
depth passed its seasonality test is
flagged 3.
Cup that failed the seasonality test, and who’s
counterpart at the next depth also failed its
seasonality test is flagged 2.
QC flag 1
Figure 6: %PIC flux QC test sequence and results.
18 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
Test conditions:
Cup %PIC seasonality test
Table 5. %PIC flux QC thresholds - 1000m
1000m (1997 2017, 47°S)
Time of integration
Average %PIC ± 1 std
Lower limit (average 3
std)
Upper limit (average + 3
std)
21st June 4th July
7.1 ± 1.4
3.0
11.2
5th July 18th July
7.1 ± 1.4
2.8
11.3
19th July 1st Aug
7.1 ± 1.4
2.9
11.3
2nd Aug 15th Aug
7.2 ± 1.0
4.2
10.2
16th Aug 29th Aug
7.4 ± 0.9
4.6
10.2
30th Aug 12th Sept
7.6 ± 1.0
4.8
10.5
13th Sept 26th Sept
7.8 ± 0.7
5.7
9.8
27th Sept 10th Oct
8.1 ± 0.4
6.9
9.3
11th Oct 24th Oct
8.3 ± 0.3
7.5
9.1
25th Oct 7th Nov
8.4 ± 0.4
7.2
9.6
8th Nov 21st Nov
8.2 ± 0.6
6.3
10.0
22nd Nov 5th Dec
7.5 ± 0.8
5.0
10.0
6th Dec 19th Dec
6.9 ± 1.4
2.8
11.1
20th Dec 2nd Jan
6.3 ± 1.3
2.4
10.3
3rd Jan 16th Jan
6.7 ± 0.9
3.9
9.5
17th Jan 30th Jan
6.9 ± 1.1
3.7
10.2
31st Jan 13th Feb
7.0 ± 1.6
2.3
11.7
14th Feb 27th Feb
7.5 ± 1.5
3.1
11.9
28th Feb 13th Mar
7.5 ± 1.3
3.4
11.5
14th Mar 27th Mar
7.4 ± 1.2
3.7
11.1
28th Mar 10th Apr
7.4 ± 1.2
3.8
11.0
11th Apr 24th Apr
7.7 ± 1.1
4.4
11.0
25th Apr 8th May
7.9 ± 1.0
4.8
11.0
9th May 22nd May
7.8 ± 1.2
4.3
11.2
23rd May 5th June
7.6 ± 1.0
4.6
10.7
6th June 20th June
7.4 ± 0.9
4.6
10.2
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 19
Table 6.%PIC flux QC thresholds - 2000m
2000m (1997 2017, 47°S)
Time of integration
Average %PIC ± 1 std
Lower limit (average 3
std)
Upper limit (average + 3
std)
21st June 4th July
8.1 ± 1.0
5.2
11.0
5th July 18th July
8.1 ± 1.0
5.1
11.0
19th July 1st Aug
8.1 ± 0.8
5.7
10.5
2nd Aug 15th Aug
8.2 ± 0.6
6.2
10.1
16th Aug 29th Aug
8.1 ± 0.5
6.6
9.6
30th Aug 12th Sept
8.0 ± 0.4
6.7
9.2
13th Sept 26th Sept
8.1 ± 0.3
7.1
9.1
27th Sept 10th Oct
8.2 ± 0.2
7.5
9.0
11th Oct 24th Oct
8.4 ± 0.2
7.9
8.9
25th Oct 7th Nov
8.6 ± 0.3
7.6
9.6
8th Nov 21st Nov
8.5 ± 0.7
6.5
10.5
22nd Nov 5th Dec
8.0 ± 0.9
5.1
10.8
6th Dec 19th Dec
7.3 ± 1.3
3.4
11.1
20th Dec 2nd Jan
7.1 ± 1.2
3.5
10.7
3rd Jan 16th Jan
7.4 ± 0.9
4.6
10.1
17th Jan 30th Jan
7.6 ± 0.9
5.0
10.2
31st Jan 13th Feb
7.6 ± 0.7
5.6
9.7
14th Feb 27th Feb
7.9 ± 0.7
5.9
9.9
28th Feb 13th Mar
8.0 ± 0.7
5.8
10.2
14th Mar 27th Mar
8.0 ± 0.7
5.8
10.2
28th Mar 10th Apr
8.1 ± 0.7
6.0
10.3
11th Apr 24th Apr
8.1 ± 0.8
5.8
10.4
25th Apr 8th May
8.1 ± 0.7
6.1
10.1
9th May 22nd May
7.9 ± 0.4
6.5
9.2
23rd May 5th June
7.8 ± 0.7
5.6
10.1
6th June 20th June
7.9 ± 1.4
3.8
12.0
20 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
Table 7. %PIC flux QC thresholds - 3800m
3800m (1997 2017, 47°S)
Time of integration
Average %PIC ± 1 std
Lower limit (average 3
std)
Upper limit (average + 3
std)
21st June 4th July
8.6 ± 0.6
6.7
10.4
5th July 18th July
8.7 ± 0.5
7.2
10.3
19th July 1st Aug
8.7 ± 0.5
7.2
10.3
2nd Aug 15th Aug
8.7 ± 0.4
7.4
10.0
16th Aug 29th Aug
8.7 ± 0.4
7.6
9.9
30th Aug 12th Sept
8.7 ± 0.4
7.6
9.9
13th Sept 26th Sept
8.8 ± 0.3
8.0
9.6
27th Sept 10th Oct
8.9 ± 0.2
8.2
9.6
11th Oct 24th Oct
8.8 ± 0.2
8.3
9.4
25th Oct 7th Nov
8.7 ± 0.4
7.5
9.9
8th Nov 21st Nov
8.7 ± 0.5
7.3
10.1
22nd Nov 5th Dec
8.5 ± 0.6
6.8
10.3
6th Dec 19th Dec
8.2 ± 0.9
5.4
11.0
20th Dec 2nd Jan
7.8 ± 1.0
4.8
10.8
3rd Jan 16th Jan
7.8 ± 0.9
5.1
10.5
17th Jan 30th Jan
7.6 ± 0.9
5.0
10.3
31st Jan 13th Feb
7.5 ± 1.0
4.5
10.6
14th Feb 27th Feb
7.8 ± 0.8
5.4
10.2
28th Feb 13th Mar
8.1 ± 0.6
6.3
9.9
14th Mar 27th Mar
8.2 ± 0.6
6.5
10.0
28th Mar 10th Apr
8.3 ± 0.6
6.4
10.2
11th Apr 24th Apr
8.4 ± 0.7
6.2
10.5
25th Apr 8th May
8.5 ± 0.8
6.1
10.9
9th May 22nd May
8.5 ± 0.8
6.1
10.9
23rd May 5th June
8.4 ± 0.9
5.8
11.0
6th June 20th June
8.3 ± 0.8
5.9
10.8
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 21
ESTIMATION OF ANALYTICAL UNCERTAINTIES
Analytical measurement uncertainty is calculated by combining accuracy and precision estimates.
In the following examples of the calculations are given along with the long-term average expanded
measurement uncertainty.
Accuracy is assessed with the help of calibration standard recoveries, CaCO3, and calculated as
follows:
(9)   


Example:
CaCO3
% recovery
Standard
deviation of %
recoveries
Number of CaCO3
standards
Mean %
recovery
CaCO3 Std.1 283.8µg
98.64
0.52
6
99.79
CaCO3 Std.2 989.9 µg
99.90
CaCO3 Std.3 2018.5 µg
100.16
CaCO3 Std.4 2845.5 µg
99.97
CaCO3 Std.5 3773.1 µg
99.98
CaCO3 Std.3 2156.4 µg
100.10
Relative standard uncertainty of analysis accuracy = (0.52 / (6)) / 99.79 = 0.002 or 0.21%
Precision is assessed with the help of sample duplicates, and calculated as follows:
(10)    
 
22 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
Example:
Sample duplicates
%CaCO3
(Difference
between
duplicates)/mean
47_3800_3a
66.28
0.007
47_3800_3b
65.81
47_2000_3a
65.71
-0.011
47_2000_3b
66.44
47_2000_17a
76.87
-0.008
47_2000_17b
77.49
47_1000_9a
65.32
-0.007
47_1000_9b
65.80
47_1000_14b
73.44
0.021
47_1000_14b
71.91
std
0.012
Relative standard uncertainty an analysis precision (%CaCO3) = 0.009 or 0.85%
The combined measurement uncertainty is calculated as follows:
(11) 
Example:
%CaCO3
(0.0022 + 0.00862) = 0.009 or 0.88%
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 23
The combined measurement uncertainty is reported as expanded uncertainty with a coverage
factor of 2, to represent a 95% level of confidence.
(12)  
Example:
%CaCO3 0.009 *2 = 0.018 or 1.75%
(Ellison and Williams 2012, Rees et al. 2018)
The long-term expanded measurement uncertainty for
%CaCO3 is ± 0.02 or 1.9%
%PIC fluxes are calculated from %CaCO3 results as follows:
(13)    
24 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
3.2.4 CHN Elemental Analysis
QC flags for component fluxes are based on the quality of the chemical analyses and then subjected
to QC tests similar to those outlined for mass fluxes. The quality of the chemical analysis will only
overrule the QC flag based on QC tests if the quality of the chemical analysis was compromised and
the sample could not be re-run.
ANALYTICAL QUALITY CONT ROL TESTS
Analytical runs contain reference materials, PACS-2 (a sediment) and Acetanilide, a certified
reference material. %PC and %PN results of these reference materials are recorded in a quality
control chart of results over time.
Upper and lower quality control limits for reference material %PC (w/w) and %PN (w/w) results
were set to 2 times standard deviation.
About 10% of all collected samples are usually run in duplicates. Where available, duplicate sample
results are recorded in a control chart. The percent difference between duplicates from their mean
is plotted over time. The POC/PN ratio of samples was compared to seasonally averaged time-series
results, with the upper and lower control limits set to 3 times standard deviation.
If either i) more than one reference material result fell outside the control limits, ii) the POC/PN
ratio fell outside of control limits or, iii) if the relative percent difference of duplicates was more
than 5% (Conte et al. 2001, Martiny et al. 2014), the affected duplicates and a subset of samples of
the affected run are repeated. If the repeated duplicates or the repeated single sample results
deviate more than 5% from previous results, all previous samples are re-run. Otherwise results of
the new set of duplicates and the average of the repeated and previous single results are accepted
and the data is flagged as 1.
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 25
Test conditions:
POC/PN seasonality test
Table 8. POC/PN flux QC thresholds - 1000m
1000m (1997 2017, 47°S)
Time interval
Average POC / PN ± 1 std
Lower limit (average 3
std)
Upper limit (average + 3
std)
21st June 4th July
7.1 ± 1.0
4.0
10.2
5th July 18th July
7.1 ± 1.2
3.5
10.7
19th July 1st Aug
7.3 ± 0.8
4.8
9.8
2nd Aug 15th Aug
7.1 ± 0.3
6.2
8.1
16th Aug 29th Aug
7.0 ± 0.5
5.6
8.4
30th Aug 12th Sept
6.7 ± 0.4
5.4
8.0
13th Sept 26th Sept
6.8 ± 0.5
5.5
8.2
27th Sept 10th Oct
7.2 ± 0.5
5.5
8.8
11th Oct 24th Oct
7.3 ± 0.6
5.5
9.1
25th Oct 7th Nov
7.4 ± 0.7
5.4
9.5
8th Nov 21st Nov
7.3 ± 1.0
4.3
10.3
22nd Nov 5th Dec
7.8 ± 1.3
4.0
11.5
6th Dec 19th Dec
7.3 ± 1.3
3.3
11.3
20th Dec 2nd Jan
7.3 ± 1.3
3.3
11.4
3rd Jan 16th Jan
7.4 ± 1.4
3.1
11.8
17th Jan 30th Jan
7.4 ± 1.2
3.7
11.1
31st Jan 13th Feb
7.3 ± 0.9
4.6
10.0
14th Feb 27th Feb
7.4 ± 1.0
4.4
10.4
28th Feb 13th Mar
7.2 ± 0.9
4.4
10.0
14th Mar 27th Mar
7.1 ± 1.2
3.4
10.8
28th Mar 10th Apr
6.9 ± 1.6
2.2
11.6
11th Apr 24th Apr
7.0 ± 1.2
3.4
10.5
25th Apr 8th May
7.0 ± 1.0
3.9
10.0
9th May 22nd May
6.6 ± 1.1
3.3
9.8
23rd May 5th June
6.9 ± 0.6
5.0
8.8
6th June 20th June
7.2 ± 0.5
5.9
8.6
26 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
Table 9. POC/PN flux QC thresholds - 2000m
2000m (1997 2017, 47°S)
Time interval
Average POC / PN ± 1 std
Lower limit (average 3
std)
Upper limit (average + 3
std)
21st June 4th July
7.9 ± 0.7
5.8
10.0
5th July 18th July
7.9 ± 0.7
5.7
10.1
19th July 1st Aug
7.8 ± 0.6
5.9
9.7
2nd Aug 15th Aug
7.7 ± 0.5
6.1
9.4
16th Aug 29th Aug
7.7 ± 0.6
6.0
9.5
30th Aug 12th Sept
7.5 ± 0.8
5.2
9.8
13th Sept 26th Sept
7.5 ± 0.7
5.3
9.7
27th Sept 10th Oct
7.5 ± 0.5
5.9
9.2
11th Oct 24th Oct
7.9 ± 0.4
6.7
9.1
25th Oct 7th Nov
8.0 ± 0.6
6.3
9.7
8th Nov 21st Nov
8.2 ± 0.6
6.6
9.9
22nd Nov 5th Dec
8.5 ± 0.5
7.0
10.1
6th Dec 19th Dec
8.3 ± 0.6
6.6
9.9
20th Dec 2nd Jan
7.6 ± 1.7
2.6
12.6
3rd Jan 16th Jan
8.0 ± 0.4
7.0
9.1
17th Jan 30th Jan
8.1 ± 0.3
7.2
9.0
31st Jan 13th Feb
7.8 ± 0.3
6.8
8.9
14th Feb 27th Feb
7.9 ± 0.9
5.3
10.5
28th Feb 13th Mar
7.7 ± 0.8
5.4
10.1
14th Mar 27th Mar
7.8 ± 0.7
5.7
9.9
28th Mar 10th Apr
7.8 ± 0.5
6.3
9.3
11th Apr 24th Apr
7.8 ± 0.5
6.2
9.3
25th Apr 8th May
7.7 ± 0.6
5.9
9.4
9th May 22nd May
7.8 ± 0.4
6.4
9.1
23rd May 5th June
7.8 ± 0.8
5.4
10.2
6th June 20th June
7.8 ± 0.7
5.8
9.8
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 27
Table 10. POC/PN flux QC thresholds - 3800m
3800m (1997 2017, 47°S)
Time interval
Average POC / PN ± 1 std
Lower limit (average 3
std)
Upper limit (average + 3
std)
21st June 4th July
7.9 ± 0.8
5.5
10.3
5th July 18th July
7.9 ± 0.8
5.5
10.4
19th July 1st Aug
7.8 ± 1.0
4.8
10.7
2nd Aug 15th Aug
7.5 ± 1.2
3.9
11.1
16th Aug 29th Aug
7.8 ± 1.0
4.8
10.8
30th Aug 12th Sept
8.0 ± 0.8
5.7
10.3
13th Sept 26th Sept
8.2 ± 1.0
5.2
11.2
27th Sept 10th Oct
8.0 ± 1.1
4.7
11.4
11th Oct 24th Oct
8.1 ± 1.0
5.1
11.2
25th Oct 7th Nov
8.7 ± 1.2
5.1
12.2
8th Nov 21st Nov
8.5 ± 1.2
4.8
12.1
22nd Nov 5th Dec
8.5 ± 0.9
5.7
11.2
6th Dec 19th Dec
8.2 ± 0.9
5.4
11.1
20th Dec 2nd Jan
8.2 ± 1.1
4.8
11.7
3rd Jan 16th Jan
8.2 ± 1.0
5.2
11.1
17th Jan 30th Jan
7.9 ± 0.8
5.6
10.2
31st Jan 13th Feb
8.0 ± 0.7
6.0
10.0
14th Feb 27th Feb
8.1 ± 0.7
6.1
10.1
28th Feb 13th Mar
8.5 ± 0.7
6.4
10.5
14th Mar 27th Mar
8.7 ± 0.8
6.4
11.1
28th Mar 10th Apr
8.6 ± 0.9
6.0
11.3
11th Apr 24th Apr
8.4 ± 0.9
5.6
11.1
25th Apr 8th May
8.3 ± 0.9
5.5
11.1
9th May 22nd May
8.2 ± 1.0
5.2
11.2
23rd May 5th June
8.2 ± 1.1
4.8
11.6
6th June 20th June
8.3 ± 1.1
4.9
11.7
28 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
Control charts:
Figure 7: %C (top) and %N (bottom) of reference material PACS-2 over time as chemical analysis control chart. Solid
line = mean, dash-dotted line = 2 SD.
2.6
2.8
3.0
3.2
3.4
3.6
3.8
4.0
4.2
4.4
010 20 30 40 50 60
%C PACS-2
Running number
PACS-2 reference material
0.22
0.24
0.26
0.28
0.30
0.32
0.34
0.36
010 20 30 40 50 60
%N PACS-2
Running number
PACS-2 reference material
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 29
Figure 8: %C (top) and %N (bottom) of reference material Acetanilide over time as chemical analysis control chart.
Solid line = mean, dash-dotted line = 2 SD.
60
65
70
75
80
85
010 20 30 40 50 60
%C Acetanilide
Running number
Acetanilide reference material
8
9
9
10
10
11
11
12
12
13
010 20 30 40 50 60
%N Acetanilide
Running number
Acetanilide reference material
30 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
Figure 9: %C (top) and %N (bottom) of duplicates over time as chemical analysis control chart. Dash-dotted line = 5%.
-6.0%
-4.0%
-2.0%
0.0%
2.0%
4.0%
6.0%
020 40 60 80 100 120 140
%PC duplicates
-20%
-10%
0%
10%
20%
30%
40%
50%
60%
020 40 60 80 100 120 140
%PN duplicates
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 31
SAMPLE RESULT QC TESTS
After quality of analytical results has been assured, particle composition values are subjected to
quality control tests similar to those described for mass fluxes based on time series averages and
3 standard deviations as upper and lower limit (Fig 10 and 11).
Cups that have failed QC tests are subjected to closer scrutiny. For this purpose, processing notes
and sample photos are consulted. Special conditions of each sample, such as large quantities of soft
bodied >1mm organisms that may break up during sieving can contribute disproportionately to the
mass and POC/PN ratios. Foul smell of the cup’s content on recovery can indicate bacterial
respiration and potentially affect POC/PN ratios. If cups fail all QC tests and processing notes have
indicated potentially compromising conditions, then the final QC flag will be 4.
Very high concentrations of calcifying organisms, often visible as white sand on the <1mm fraction
filters, will lead to high %PIC and thus high %PC results. These are situations that can explain
unexpected results that will lead to QC test failures but will not lead to a high QC flag.
i. Cup CHN seasonality test
ii. Cup-to-cup CHN QC comparison across depth
iii. Cup %PIC seasonality test
Cup seasonality %PC
Depth comparison test
Cup %PIC seasonality test
Test fail
Test pass
Cup that failed seasonality test and the
corresponding cup at the next depth also
failed its seasonality test is flagged 2.
If a cup failed its seasonality test and the corresponding cup at the
next depth passed its seasonality test, the %PIC QC test result is
used to inform the final %PC QC flag. An out of range %PIC result
can influence the %PC result. If the final %PIC QC flag is 2 or 3, then
the final PC% QC flag is matched to 2 or 3.
QC flag 1
Depth comparison passes
Depth comparison fails
If a cup failed its seasonality test and the corresponding cup
at the next depth passed its seasonality test, the %PIC QC
test result is used to inform the final %PC QC flag. An out of
range %PIC result can influence the %PC result. If the final
%PIC QC flag is 1, then the final PC% QC flag is 3.
Figure 10:%PC flux QC test sequence and results.
32 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
Cup seasonality %PN
Depth comparison test
Test fail
Test pass
Depth comparison fails
Cup that failed its seasonality test and the
corresponding cup at the next depth also
failed its seasonality test is flagged 2.
Depth comparison passes
QC flag 1
Cup that failed its seasonality test and
the corresponding cup at the next depth
passed its seasonality test is flagged 3.
Figure 11: %PN flux QC test sequence and results.
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 33
Test conditions:
Cup CHN seasonality test
Table 11. %PC flux QC thresholds - 1000m
1000m (1997 2017, 47°S)
Time interval
Average %PC ± 1 std
Lower limit (average 3
std)
Upper limit (average + 3
std)
21st June 4th July
15.7 ± 3.4
5.6
25.9
5th July 18th July
16.2 ± 3.4
5.9
26.5
19th July 1st Aug
16.4 ± 3.1
7.2
25.7
2nd Aug 15th Aug
16.1 ± 2.1
9.7
22.5
16th Aug 29th Aug
15.7 ± 1.3
11.8
19.7
30th Aug 12th Sept
15.5 ± 0.8
13.2
17.9
13th Sept 26th Sept
15.5 ± 0.7
13.3
17.8
27th Sept 10th Oct
14.9 ± 0.5
13.3
16.5
11th Oct 24th Oct
14.5 ± 0.4
13.3
15.6
25th Oct 7th Nov
14.1 ± 0.5
12.6
15.7
8th Nov 21st Nov
14.3 ± 1.3
10.5
18.1
22nd Nov 5th Dec
15.5 ± 2.8
7.2
23.8
6th Dec 19th Dec
15.0 ± 3.3
4.9
25.0
20th Dec 2nd Jan
15.6 ± 1.6
10.8
20.4
3rd Jan 16th Jan
15.7 ± 2.1
9.3
22.0
17th Jan 30th Jan
15.4 ± 2.8
7.1
23.7
31st Jan 13th Feb
15.9 ± 1.3
11.9
19.9
14th Feb 27th Feb
16.6 ± 2.0
10.5
22.6
28th Feb 13th Mar
16.2 ± 2.0
10.1
22.3
14th Mar 27th Mar
16.0 ± 2.1
9.6
22.4
28th Mar 10th Apr
16.4 ± 2.4
9.3
23.5
11th Apr 24th Apr
16.5 ± 2.6
8.6
24.4
25th Apr 8th May
16.0 ± 2.3
9.0
23.0
9th May 22nd May
16.1 ± 1.3
12.2
20.0
23rd May 5th June
16.1 ± 1.4
11.9
20.3
6th June 20th June
16.1 ± 2.1
9.8
22.4
34 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
Negative lower limits were replaced by zeros.
Table 12. %PN flux QC thresholds - 1000m
1000m (1997 2017, 47°S)
Time interval
Average %PN ± 1 std
Lower limit (average 3
std)
Upper limit (average + 3
std)
21st June 4th July
1.4 ± 0.5
0.0
2.9
5th July 18th July
1.5 ± 0.5
0.0
2.9
19th July 1st Aug
1.5 ± 0.5
0.0
2.9
2nd Aug 15th Aug
1.4 ± 0.4
0.2
2.7
16th Aug 29th Aug
1.4 ± 0.3
0.4
2.3
30th Aug 12th Sept
1.4 ± 0.4
0.3
2.5
13th Sept 26th Sept
1.3 ± 0.2
0.8
1.8
27th Sept 10th Oct
1.1 ± 0.1
0.8
1.5
11th Oct 24th Oct
1.0 ± 0.1
0.7
1.3
25th Oct 7th Nov
0.9 ± 0.1
0.6
1.2
8th Nov 21st Nov
1.0 ± 0.3
0.0
1.9
22nd Nov 5th Dec
1.2 ± 0.6
0.0
3.0
6th Dec 19th Dec
1.3 ± 0.8
0.0
3.7
20th Dec 2nd Jan
1.3 ± 0.4
0.0
2.6
3rd Jan 16th Jan
1.4 ± 0.4
0.1
2.7
17th Jan 30th Jan
1.3 ± 0.5
0.0
2.6
31st Jan 13th Feb
1.3 ± 0.4
0.2
2.5
14th Feb 27th Feb
1.4 ± 0.6
0.0
3.2
28th Feb 13th Mar
1.4 ± 0.6
0.0
3.2
14th Mar 27th Mar
1.4 ± 0.6
0.0
3.2
28th Mar 10th Apr
1.5 ± 0.7
0.0
3.5
11th Apr 24th Apr
1.4 ± 0.7
0.0
3.5
25th Apr 8th May
1.3 ± 0.6
0.0
3.1
9th May 22nd May
1.4 ± 0.3
0.5
2.3
23rd May 5th June
1.3 ± 0.3
0.5
2.2
6th June 20th June
1.4 ± 0.3
0.4
2.3
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 35
Table 13. %PC flux QC thresholds - 2000m
2000m (1997 2017, 47°S)
Time interval
Average %PC ± 1 std
Lower limit (average 3
std)
Upper limit (average + 3
std)
21st June 4th July
14.7 ± 2.0
8.7
20.8
5th July 18th July
14.6 ± 2.1
8.3
20.9
19th July 1st Aug
14.3 ± 1.5
9.8
18.8
2nd Aug 15th Aug
14.1 ± 0.9
11.4
16.7
16th Aug 29th Aug
14.0 ± 0.4
12.8
15.2
30th Aug 12th Sept
14.1 ± 0.8
11.7
16.5
13th Sept 26th Sept
14.1 ± 1.2
10.5
17.6
27th Sept 10th Oct
14.2 ± 1.1
10.9
17.5
11th Oct 24th Oct
14.0 ± 1.1
10.9
17.2
25th Oct 7th Nov
13.7 ± 0.7
11.6
15.9
8th Nov 21st Nov
13.2 ± 0.4
11.9
14.6
22nd Nov 5th Dec
13.3 ± 0.5
11.7
14.9
6th Dec 19th Dec
13.5 ± 0.7
11.5
15.6
20th Dec 2nd Jan
13.4 ± 1.0
10.5
16.4
3rd Jan 16th Jan
13.3 ± 0.5
11.9
14.6
17th Jan 30th Jan
13.5 ± 0.7
11.5
15.5
31st Jan 13th Feb
14.8 ± 1.6
10.2
19.5
14th Feb 27th Feb
14.3 ± 1.3
10.3
18.3
28th Feb 13th Mar
13.9 ± 1.3
10.1
17.8
14th Mar 27th Mar
14.0 ± 1.6
9.3
18.8
28th Mar 10th Apr
14.1 ± 1.3
10.0
18.1
11th Apr 24th Apr
14.3 ± 1.3
10.4
18.1
25th Apr 8th May
14.4 ± 1.3
10.5
18.2
9th May 22nd May
15.3 ± 1.6
10.3
20.2
23rd May 5th June
15.7 ± 3.2
6.0
25.3
6th June 20th June
15.9 ± 5.2
0.3
31.5
36 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
Negative lower limits were replaced by zeros.
Table 14. %PN flux QC thresholds - 2000m
2000m (1997 2017, 47°S)
Time interval
Average %PN ± 1 std
Lower limit (average 3
std)
Upper limit (average + 3
std)
21st June 4th July
1.0 ± 0.5
0
2.3
5th July 18th July
1.0 ± 0.5
0
2.3
19th July 1st Aug
0.9 ± 0.4
0
2.0
2nd Aug 15th Aug
0.9 ± 0.2
0.2
1.6
16th Aug 29th Aug
0.9 ± 0.1
0.5
1.3
30th Aug 12th Sept
1.0 ± 0.2
0.3
1.6
13th Sept 26th Sept
0.9 ± 0.3
0.1
1.8
27th Sept 10th Oct
0.9 ± 0.3
0.2
1.7
11th Oct 24th Oct
0.8 ± 0.2
0.2
1.5
25th Oct 7th Nov
0.7 ± 0.2
0.2
1.3
8th Nov 21st Nov
0.7 ± 0.1
0.2
1.1
22nd Nov 5th Dec
0.7 ± 0.2
0.1
1.3
6th Dec 19th Dec
0.9 ± 0.2
0.2
1.6
20th Dec 2nd Jan
0.9 ± 0.3
0
1.9
3rd Jan 16th Jan
0.9 ± 0.2
0.4
1.3
17th Jan 30th Jan
0.9 ± 0.1
0.4
1.3
31st Jan 13th Feb
1.1 ± 0.3
0.2
1.9
14th Feb 27th Feb
0.9 ± 0.2
0.3
1.6
28th Feb 13th Mar
0.9 ± 0.2
0.2
1.6
14th Mar 27th Mar
0.9 ± 0.3
0.1
1.7
28th Mar 10th Apr
0.9 ± 0.3
0.1
1.7
11th Apr 24th Apr
0.9 ± 0.3
0.1
1.7
25th Apr 8th May
0.9 ± 0.3
0.2
1.7
9th May 22nd May
1.1 ± 0.3
0.1
2.1
23rd May 5th June
1.2 ± 0.6
0
3.0
6th June 20th June
1.2 ± 0.9
0
4.0
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 37
Table 15. %PC flux QC thresholds - 3800m
3800m (1997 2017, 47°S)
Time interval
Average %PC ± 1 std
Lower limit (average 3
std)
Upper limit (average + 3
std)
21st June 4th July
13.0 ± 1.0
10.0
16.0
5th July 18th July
12.9 ± 0.9
10.2
15.6
19th July 1st Aug
12.9 ± 1.0
9.7
16.0
2nd Aug 15th Aug
12.9 ± 0.9
10.2
15.6
16th Aug 29th Aug
12.9 ± 0.7
10.7
15.1
30th Aug 12th Sept
12.8 ± 0.6
10.9
14.8
13th Sept 26th Sept
12.5 ± 0.4
11.2
13.8
27th Sept 10th Oct
12.5 ± 0.4
11.3
13.7
11th Oct 24th Oct
12.7 ± 0.3
11.6
13.7
25th Oct 7th Nov
12.9 ± 0.5
11.3
14.5
8th Nov 21st Nov
12.7 ± 0.4
11.5
13.8
22nd Nov 5th Dec
12.8 ± 0.5
11.2
14.4
6th Dec 19th Dec
12.8 ± 0.6
10.9
14.6
20th Dec 2nd Jan
12.8 ± 0.5
11.2
14.4
3rd Jan 16th Jan
12.7 ± 0.4
11.4
14.0
17th Jan 30th Jan
12.9 ± 0.7
10.7
15.0
31st Jan 13th Feb
13.1 ± 0.9
10.4
15.8
14th Feb 27th Feb
13.3 ± 0.9
10.6
16.0
28th Feb 13th Mar
13.4 ± 1.0
10.5
16.4
14th Mar 27th Mar
13.5 ± 1.1
10.1
16.9
28th Mar 10th Apr
13.5 ± 1.2
10.0
16.9
11th Apr 24th Apr
13.3 ± 0.8
10.7
15.8
25th Apr 8th May
13.2 ± 0.6
11.3
15.1
9th May 22nd May
13.2 ± 0.5
11.6
14.7
23rd May 5th June
13.2 ± 0.8
10.7
15.8
6th June 20th June
13.3 ± 1.0
10.3
16.3
38 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
Negative lower limits were replaced by zeros.
Table 16. %PN flux QC thresholds - 3800m
3800m (1997 2017, 47°S)
Time interval
Average %PN ± 1 std
Lower limit (average 3
std)
Upper limit (average + 3
std)
21st June 4th July
0.7 ± 0.2
0.0
1.3
5th July 18th July
0.6 ± 0.2
0.0
1.2
19th July 1st Aug
0.6 ± 0.2
0.0
1.2
2nd Aug 15th Aug
0.7 ± 0.2
0.0
1.3
16th Aug 29th Aug
0.6 ± 0.2
0.1
1.1
30th Aug 12th Sept
0.6 ± 0.1
0.3
0.9
13th Sept 26th Sept
0.5 ± 0.1
0.3
0.7
27th Sept 10th Oct
0.5 ± 0.1
0.3
0.8
11th Oct 24th Oct
0.6 ± 0.1
0.3
0.8
25th Oct 7th Nov
0.6 ± 0.1
0.3
0.9
8th Nov 21st Nov
0.5 ± 0.1
0.2
0.9
22nd Nov 5th Dec
0.6 ± 0.2
0.1
1.0
6th Dec 19th Dec
0.6 ± 0.2
0.0
1.3
20th Dec 2nd Jan
0.7 ± 0.2
0.0
1.4
3rd Jan 16th Jan
0.7 ± 0.2
0.1
1.3
17th Jan 30th Jan
0.8 ± 0.2
0.2
1.4
31st Jan 13th Feb
0.8 ± 0.2
0.2
1.4
14th Feb 27th Feb
0.8 ± 0.2
0.3
1.3
28th Feb 13th Mar
0.7 ± 0.2
0.2
1.2
14th Mar 27th Mar
0.7 ± 0.2
0.2
1.2
28th Mar 10th Apr
0.7 ± 0.2
0.1
1.3
11th Apr 24th Apr
0.7 ± 0.2
0.1
1.2
25th Apr 8th May
0.7 ± 0.2
0.1
1.2
9th May 22nd May
0.7 ± 0.2
0.1
1.3
23rd May 5th June
0.7 ± 0.3
0
1.5
6th June 20th June
0.7 ± 0.3
0
1.5
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 39
ESTIMATION OF ANALYTICAL UNCERTAINTIES
Analytical measurement uncertainty is calculated by combining accuracy and precision estimates.
In the following examples of the calculations are given along with the long-term average expanded
measurement uncertainty.
Accuracy is assessed with the help of certified reference material, Acetanilide, and calculated as
follows:
(5) 
 
 
 
Example:
Acetanilide reference value %C = 71.05 %N = 10.42
Acetanilide results
%C
(Sample result /
certified
value)*100 = %
recovery
%N
(Sample result /
certified
value)*100 = %
recovery
# 1
71.14
100.12
10.35
99.34
# 2
70.94
99.85
10.26
98.48
# 3
70.99
99.92
10.30
98.81
# 4
70.93
99.83
10.21
97.94
# 5
70.92
99.82
10.27
98.55
# 6
71.11
100.09
10.32
99.04
mean
99.94
mean
std
0.12
std
std (%C recovery) / mean (%C recovery) / (6) = 0.0005
std (%N recovery) / mean (%N recovery) / (6) = 0.0018
40 | Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report.
Sediment trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d)
relative standard uncertainty of analysis accuracy (%C) = 0.0005 or 0.05%
relative standard uncertainty of analysis accuracy (%N) = 0.0018 or 0.18%
Precision is assessed with the help of sample duplicates, and calculated as follows:
(6)   
 
Example:
Sample duplicates
%C
(Difference
between
duplicates)/mean
%N
(Difference
between
duplicates)/mean
47_1000_9a
17.051
0.005
1.690
-0.016
47_1000_9b
16.973
1.717
47_1000_14a
14.712
0.010
0.969
0.014
47_1000_14b
14.573
0.955
47_2000_3a
16.675
-0.005
1.433
-0.015
47_2000_3b
16.751
1.454
47_2000_17a
12.733
-0.032
0.551
-0.040
47_2000_17b
13.151
0.573
47_3800_3a
15.902
-0.002
1.197
0.002
47_3800_3b
15.928
1.195
std
0.015
std
0.018
relative standard uncertainty of analysis precision (%C) = 0.010 or 1.03%
relative standard uncertainty of analysis precision (%N) = 0.013 or 1.29%
Wynn-Edwards, CA, Davies, DM, Shadwick, EH, Trull, TW (2020) Southern Ocean Time Series. SOTS Quality assessment and control report. Sediment
trap particle fluxes Version 1.0. CSIRO, Australia. DOI: 10.26198/5dfad21358a8d (http://dx.doi.org/10.26198/5dfad21358a8d) 41
The combined measurement uncertainty is calculated as follows:
(7) 
Example:
%C
(0.00052 + 0.0112) = 0.010 or 1.03%
%N
(0.00182 + 0.0132) = 0.013 or 1.31%
The combined measurement uncertainty is reported as expanded uncertainty with a coverage
factor of 2, to represent a 95% level of confidence.
(8)  
Example:
%C 0.010 *2 = 0.021 or 2.06%
%N 0.013 *2 = 0.026 or 2.62%
(Ellison and Williams 2012, Rees et al. 2018)
The long-term expanded measurement uncertainty for
%C is ± 0.02 or 2.1%
%N is ± 0.04 or 3.8%