1. The IQuOD initiative International Quality-Controlled Ocean Database Current international partners: Argentina, Australia, Brazil, Canada, China, France, Germany, India, Japan, Mexico, Norway, Russia, Spain, South Africa, UK, USA. Co-Chairs: Catia Domingues, ACE CRC, Australia Rebecca Cowley, CSIRO Oceans and Atmosphere Australia Steering Team: Tim Boyer (NOAA), Ann Thresher (CSIRO), Simon Good (UK Met Office), Matt Palmer (UK Met Office), Susan Wijffels (CSIRO), Gustavo Goni (AOML), Janet Sprintall (Scripps), Alison Macdonald (WHOI), Toru Suzuki (MIRC), Viktor Gouretski (ICDC), Steve Diggs (UCSD)

2. Mission statement IQuOD aims to produce, freely distribute and curate the highest quality, most complete and consistent global subsurface ocean temperature* profile database possible. IQuOD will include intelligent metadata and assign an estimated uncertainty to each individual observation. What is IQuOD about? The power of IQuOD: Ability to pull together the expertise from the international research community (producers/users) and to focus that combined effort into a single “best” dataset.

3. What are the motivators for IQuOD? Societal needs Understanding climate variability and Put modern changes in the context of past changes (e.g., mean trends, extremes) Separate anthropogenic drivers from natural climate modes of variability (e.g., ENSO, NAO, IOD, SAM, PDO, AMO, etc). Evaluate, constrain, initialize, assimilate into numerical models. Earth’s total heat content >90% Rhein et al. (2013)

4. Today’s big challenge: ‘Climate quality’ ocean database No. Profiles / year Courtesy of Tim Boyer, IQuOD 2 nd Workshop, June 2014

5. The contribution of XBT errors to ocean warming estimates: Warming trend, 0- 700m Abraham et al, 2014 Latitude Temperature bias (°C) Year Biases derived from Cowley et al, 2013

6. Manual QC reveals warm biases Gaussian Warm temperature bias remains Gronell and Wijffels (2008) Pilot test: Indian Ocean/SW Pacific Josh Willis and Ann Thresher

7. Measurements vs. Depth WOD13 Depth (m) Green – Temperature Red – Salinity Yellow – Silicate Pink - Chlorophyll Courtesy of Tim Boyer, IQuOD 2 nd Workshop, June 2014 Yes CTD XBT/XCTD Bottle MBT Bucket Unknown No Argo Glider Drifters TSG Maybe Pinniped Moored buoy Others? Why focus on temperature?

8. QC experts (automated/manual) Data management experts Historical/modern in situ instrumentation experts Regional oceanography experts Program managers Ocean Reanalyses experts (users) Climate model experts (users) Who? International collaboration/cooperation is the key to success

9. How will IQuOD work? Co-Chairs: Catia Domingues, Rebecca Cowley Steering committee Tim Boyer, Ann Thresher, Simon Good, Matt Palmer, Susan Wijffels, Gustavo Goni, Janet Sprintall, Alison Macdonald, Toru Suzuki, Steve Diggs, Viktor Gouretski Task Team 1 GDAC (Tim Boyer) Task Team 2 Auto QC (Bec Cowley) Task Team 4 Uncertainties/Formats (Simon Good) Task Team 3 Manual QC (Ann Thresher) Task Team 5 Aggregation (Catia Domingues)

10. Instrument type information OSDCTDPFLAPBMBTXBTMRB Depth Range, km 0-8 0-2 0-0.30-1.90-4 Measurement uncertainty for T (random error), degr. C ~0.02-0.1 ~0.001-0.01 ~0.005~0.05~0.2 ? Measurement uncertainty for Sample Depth (random error), m 1-5%< 0.015% ? Sensor drift? ? Sensor drift? 1-3%2%(5m)? Temp. Bias, degr. C ????O(0.1) ? Depth Bias, m ? ? ??1-5% ? Metadata Availability poor?Good? ?Very poor Poor? Manufacturers ?????2 (4)?

11. Summary IQuOD will: Provide a unique ocean profile dataset essential for climate applications and model validations Goals for this workshop (from the IQuOD point of view): The uncertainties associated with XBT data Before and after corrections are applied A statement of what the biases are, their size & uncertainty Recommended metadata for collection with XBT data A recommended correction for historical XBT data A statement of future goals for XBT bias corrections

12. Thank you

13. What makes IQuOD unique? Dataset Automated QC Manual/Visual QC (~40% of data) $$ citizen science? IQuOD dataset available (version control) Software, training

14. A major problem: missing metadata & original profiles Number of known & unknown XBT profiles per year Proportion of unknown XBT profiles (shallow) Abraham et al. (2013)

15. Instrumental (time-dependent) biases – MBTs/XBTs Closer scrutinity: Gouretski and Koltermann (2007) Wijffels et al. (2008) Domingues et al. (2008) Implications for ocean warming variability and trend

16. OHC variability and change: sensitive to small systematic errors Impact of time-dependent XBT bias corrections Cowley et al. (2013) Abraham et al. (2013) Various XBT corrections proposed Boyer et al. (in progress) No correction

17. Notes Availability of data and metadata v important Open access. JCOMM has a 10-year vision to develop the MCDS (Marine climate data system) MCDS will make high quality marine data accessible. Identifying gaps in the state of marine climate data componenet of global climate observing system. Outcome will be a paper in international journal of climatology Figure with task groups and structure, some pics too. Thermocline feedback Alliances etc get established, but what work happens? IQUOD has outcomes. Added value of datasets: fishing/weather prediction Essential observations.