1. 1 Edinburgh, Scotland Kenneth S. Casey Chair, GHRSST RAN-TAG NOAA National Oceanographic Data Center 27 June – 01 July, 2011 GHRSST Reanalysis Technical Advisory Group (RAN-TAG) 2011 Breakout Session Kenneth S. Casey Chair, GHRSST RAN-TAG NOAA National Oceanographic Data Center 27 June – 01 July, 2011

2. 2 Outline 1.Membership, Outreach, Terms of Reference, Actions (goal: update and inform) 2.Review of 2011 Draft International Status Report (goal: update and inform) 3.Discussion on Data Processing Framework for the SST ECV (goal: discuss and define) 4.Priority Actions for 2011-2012 (goal: define and document) 5.New Chair Perros-Guirec, June 2008

3. 3 1. Membership Outreach Terms of Reference Actions

4. 4 RAN Technical Advisory Group Kenneth S. Casey, Chair Edward Armstrong Jorge Vazquez Craig Donlon Hiroshi Kawamura Gilles Larnicol David Llewellyn-Jones Nick Rayner Gary Wick Richard W. Reynolds Bill Emery Helen Beggs Gary Corlett Jonah Roberts-Jones Mike Chin Eileen Maturi Matthew Martin Alexey Kaplan Steinar Eastwood Tess Brandon Bruno Buongiorno Nardelli Chris Merchant New Members? Anyone wish to step down?

5. 5 RAN-TAG On the Road MARCDAT-3, Frascati Italy University of Maryland Center for Environmental Science, Horn Point MD, USA Joint GHRSST TAG Meeting, Boulder CO, USA JCOMM Pilot Project for WIGOS Meeting, Oostende Belgium NASA SST Science Team, Seattle WA, USA 17 th AMS Conference on Satellite Oceanography and Meteorology, Annapolis MD, USA

6. 6 RAN-TAG Terms of Reference Not updated since RAN_1.0.doc Reviewed by Chair this year Need to update and simplify Document dominated by (old) requirements and (outdated) Concept of Operations New requirements at (GCOS-107 update): https://www.ghrsst.org/ghrsst- science/science-team-groups/ran-tag/latest- climate-record-requirements/ https://www.ghrsst.org/ghrsst- science/science-team-groups/ran-tag/latest- climate-record-requirements/ https://www.ghrsst.org/ghrsst- science/science-team-groups/ran-tag/latest- climate-record-requirements/ Concept of Operations and New Vision needed Update should be easy – TAKE AN ACTION

7. 7 RAN-TAG Actions ActionDescriptionTasked ToDeadlineStatus G11-8Provide more information regarding the specification of target error specs for L4 and insert into GDS-2.0 RAN-TAG (Casey)July 9 th 2010Closed: GDS2 completed. G11-48Update RAN-TAG ToR with Modern Vision and membership list. RAN-TAG (Casey)Report to G12ToR in progress, but Membership list updated and added to GHRSST web site. G11-49Publish RAN-TAG status report on GHRSST Web site RAN-TAG (Casey)Report to G12CLOSED: See https://www.ghrsst.org/d ocuments/q/category/ran -tag-documents/ G11-50Send RAN-TAG status report template to LeBorgne (MSG SEVIRI), O’Caroll (IASI), Vazquez (MODIS), and Bruce Brasnett (CMC L4) RAN-TAG (Casey)ASAPClosed G11-51Add section in GDS explaining targets for stability and bias with respect to long term in situ reference data sets. RAN-TAG (Casey)ASAPClosed with GDS2 Perros-Guirec, June 2008

8. 8 2. Review of 2011 International Status Report

9. 9 Historical and In situ Activities

10. 10 Project Goals 1.To recalibrate the AVHRR Level 1B radiances for the historic data back to 1981 2.To provide the data to the community through the NOAA CDR Program Current Status 1.Detailed analysis of MetOp-A completed showing significant improvements to the longwave channels removing a +/- 0.5K bias and showing the this AVHRR is stable to <0.08K/decade (estimated) 2.Use of instrument parameters (particularly the instrument temperature) to fix large time variable biases for NOAA- 16 3.Work beginning on systematic comparison of AVHRRs with (A)ATSR from 1991 with new hire Looking Forward 1.Analyze pre-launch data for AVHRR/1 and AVHRR/2s 2.Use multiple techniques to determine calibration for pre-1991 data 3.Use targeted SST retrieval studies as part of the validation scheme An FCDR for the AVHRR John Mittaz, NOAA/CICS Project Summary: GHRSST-12 Reanalysis Breakout Session

11. 11 HadSST3 Project Goals 1.A global, gridded data set of in situ SST measurements since 1850 with reduced biases and quantified uncertainties, including an assessment of their covariance structure Relevant Graphic Current Status 1.In press in JGR Atmospheres: Kennedy et al 2011, part 1 and 2. Available online now. 2.Data soon to be available from http://www.metoffice.gov.uk/hadobs Looking Forward 1.Currently updating from 2007 to date 2.Will soon be updated automatically every month Nick Rayner, UKMO Project Summary: GHRSST-12 Reanalysis Breakout Session

12. 12 ERSST Version 3b Project Goals 1.Provide historic SST monthly analyses from 1854 to present on a 2 o spatial grid 2.Merge SST and land surface temperatures to produce global surface temperatures on 5 o spatial grid 3.Improve documentation Current Status 1.Transferred to Global Climate Applications Division (contact: Boyin Huang) 2.Older versions not longer distributed 3.Only v3b available (no satellite inputs) AVHRR data not used because Residual Pathfinder v5 AVHRR biases impact climate signal Looking Forward 1.Reprocessing with ICOADS 2.5 2.Closer interaction with land temperature and climate assessment groups 3.New website http://www.ncdc.noaa.gov/ersst/ SST Annual Anomaly ( o C) Viva Banzon and Boyin Huang (NOAA/NCDC) Project Summary: GHRSST-12 Reanalysis Breakout Session

13. 13 Level 1 Activities

14. 14 Project Goals 1.To collect as much AVHRR HRPT data from receiving stations all over the world 2.To reformat and stitch together the collected Level 0 data, generate L1b data from that stitched data, and then produce L2 and L3 SSTs. 3.To serve as a fundamental input to high-resolution GHRSST CDRs Current Status 1.R2HA2 Working Group formed in GHRSST 2.Collected all of CMS Level 0/1 data at URI 3.SeaDAS configured to handle NOAA-archived LAC/HRPT data Looking Forward 1.Continue collection of Level 0 data at URI 2.Seek commitments from other potential partners AVHRR HRPT/LAC Reprocessing http://www.eumetsat.int Peter Cornillon, University of Rhode Island Project Summary: GHRSST-12 Reanalysis Breakout Session

15. 15 IMOS HRPT AVHRR SST Reprocessing Project Goals 1.To reprocess HRPT AVHRR data back to 1992 from Australian and Antarctic ground stations to GDS2 L2P, L3U, L3C and L3S SSTskin files 2.Aim to reduce errors by using: Regional drifting buoys to produce regression coefficients Improved BT to SST algorithms with day-time terms including latitude and higher order Improved CLAVR-based cloud clearing Nighttime NOAA-18 L3C Current Status 1.Producing RT and reprocessed HRPT AVHRR SSTskin GDS2 L2P, L3U and 1-d and 3-d day/night L3C files from NOAA-15, 17, 18 and 19 stitched raw data from Australian RX stations back to 2004 2.For 1 Nov 2009 to 31 Mar 2010, IMOS HRPT AVHRR L2P SSTfnd cf drifting buoy SSTfnd exhibited standard deviations of: NOAA-17: 0.24  C (Night), 0.35  C (Day) NOAA-18: 0.27  C (Night), 0.35  C (Day) NOAA-19: 0.26  C (Night), 0.34  C (Day) Looking Forward 1.L2P, L3U, L3C and L3S files back to 1992 available from GHRSST GDAC by Dec 2011 2.Include Casey and Davis raw AVHRR data 3.Validate AVHRR SST against IMOS in situ SST (eg. ships, Argo, seals) 4.Improve cloud/ice detection over Southern Ocean Helen Beggs, ABoM

16. 16 Optimal estimation of SST from AVHRR GAC 1978-2010 Project Goals OE SST from AVHRR GAC *Jon Mittaz/NOAA calib. ECMWF ERA atmos. RTTOV RTM CLAVR-x cloud mask. Current Status SST available 1991-2010. BTs are not yet bias-corrected, so RTM-related biases expected in SSTs. Looking Forward Correct BTs using (1)buoys (NCEO project, Merchant et al 2008) (2) ATSR (ESA SST-CCI). Extend to 1978-1991 (NCEO method). Project Summary: GHRSST-12 Reanalysis Breakout Session Filipiak, Merchant/University of Edinburgh

17. 17 Level 2 and Level 3 Activities

18. 18 NOAA NESDIS GOES EAST/WEST SSTs Project Goals Generate geostationary climate quality sea surface temperature retrievals 1994 to present (GOES-8, 9, 10, 11, 12, 13, 14, 15) Current Operational GOES-13 (GOES-E) GOES-11 (GOES-W) Current Status Improve Algorithm retrieval methodology to reprocess with improved accuracy « Implementing Physical retrieval Algorithm with aerosol correction « Improving Bayesian Cloud Mask « Calibration Correction Applied Looking Forward Seeking funding to reprocess Project Summary: GHRSST-12 Reanalysis Breakout Session

19. 19 NOAA NESDIS MT-SAT SSTs Project Goals Generate geostationary climate quality sea surface temperature retrievals for MTSAT-1 (2005-2010) MTSAT-2 (2010-present) Current Status Improve Algorithm retrieval methodology to reprocess with improved accuracy « Implementing Physical retrieval Algorithm with aerosol correction « Improving Bayesian Cloud Mask « Calibration Correction Applied Looking Forward Seeking funding to reprocess Project Summary: GHRSST-12 Reanalysis Breakout Session

20. 20 NOAA NESDIS MSG SSTs Project Goals Generate geostationary climate quality sea surface temperature retrievals from MSG-1 and MSG-2 and beyond Current Status Improve Algorithm retrieval methodology to reprocess with improved accuracy « Implementing Physical retrieval Algorithm with aerosol correction « Improving Bayesian Cloud Mask Looking Forward Seeking funding to reprocess Project Summary: GHRSST-12 Reanalysis Breakout Session

21. 21 IASI L2P SST Project Goals 1.IASI L2P SST 2.Core L2P produced from March 2010 experimentally. 3.Demonstration product available from EUMETSAT in near-real time from 24 th March 2011. Relevant Graphic Current Status 1.IASI L2 PPF upgraded to v5 in Sep10. 2.No current re-processing plans Looking Forward 1.Full IASI L2P products to be produced in the OSI-SAF CDOP-2 phase (post March 2012). 2.IASI L2 EUMETSAT products (which include surface temperature) are available from April 2008. Project Summary: GHRSST-12 Reanalysis Breakout Session Anne O’Carroll, EUMETSAT

22. 22 MODIS SST Improvements Project Goals - 1.Add LATBAND SST retrievals to MODIS Aqua and Terra 2.Implement new MODIS Collection 6 calibration 3.Minimize seasonal, latitudinal differences with respect to reference observations All Goals Achieved Current Status 1.Update to MODIS Collection 6 calibration - DONE 2.Developed new matchup databases - DONE 3.New SST retrieval equation calculated - DONE 4.New Hypercube tables being computed (discrete values) – DONE Looking Forward 1.Proposal submitted to NASA to support implementation of Hypercube with continuous functions - FUNDED 2.LATBAND and Hypercube being implemented in Pathfinder for all 5- channel AVHRRs – PENDING NOAA FUNDING? Project Summary: GHRSST-11 Reanalysis Breakout Session Bob Evans, UMiami/RSMAS Project Summary: GHRSST-12 Reanalysis Breakout Session

23. 23 AVHRR Pathfinder SST Project Goals 1.To provide the longest, most accurate, and highest resolution consistently- reprocessed SST climate data record (CDR) from the AVHRR sensor series 2.To serve as a fundamental input to GHRSST Reanalysis CDRs Current Status 1.Completed Version 5.2 GDS2.0 L3C for 1981-2010 as part of the NOAA Climate Data Record Program 2.Version 5.0 and 5.1 available: Daily, 5-, 7-, 8-day, Monthly, Yearly SSTs for 1981-2009, and climatologies 3.Included in GCOS Intercomparisons 4.Available via THREDDS, FTP, HTTP, OPeNDAP, WCS, WMS, LAS and JPL POET Looking Forward 1.Version 6 to be processed at NODC in GDS2.0 L2P, L3U, and L3C formats, with uncertainties 2.Daily, 5day and monthly V5.0 climatologies in GDS2.0 L4 format 3.HRPT/LAC “Pathfinder” coordination 4.Additional web services to be added: NODC Geoportal, ArcGIS server Pathfinder SSTs available at http://pathfinder.nodc.noaa.gov Bob Evans, RSMAS; Tess Brandon and Ken Casey, NOAA NODC Project Summary: GHRSST-12 Reanalysis Breakout Session

24. 24 NORMAP – High Latitude SST Project Goals Norwegian project for time series of satellite derived data Reprocess SST and radiative fluxes from AVHRR GAC at high latitudes (>40N) Use Climate SAF AVHRR GAC FCDR Base processing on OSI SAF Current Status - SST activity just started - Retrieving GAC, angle files and PPS cloud mask data (1989-2009 = 23Tb globally), as well as NWP from ECMWF Looking Forward - Start with 1989-2009 (currently available from CM SAF), extend with 1982-1988) - Main activity autumn 2011 and 2012 - Expect data set ready end 2012 Project Summary: GHRSST-12 Reanalysis Breakout Session Steinar Eastwood, met.no

25. 25 Project Goals 1.ATSR SST CDR 1991 – 2009 2.0.1 K accuracy; 0.05 K/dec stability; 0.1 deg resolution; homogenised over 3 sensors 3.Based on physics: independent of in situ data Relevant Graphic Current Status 1.v1.0 expected release TODAY 2.Stability assessment suggests 0.05 K / dec achieved, at least for 1993 – 2009 in tropics where it is highly testable 3.Accuracy assessment suggests 0.1 K bias target achieved globally Looking Forward 1.v1.1 to address known issues by autumn 2.Project formally has ended – work continues within SST CCI, however ATSR Reprocessing for Climate Project Summary: GHRSST-12 Reanalysis Breakout Session Chris Merchant, University of Edinburgh

26. 26 SEVIRI Project Goals 1.To reprocess SEVIRI data back to beginning of reliable reprocessed level 1 MSG-1 data, January 2004 2.The SST re-processing will be performed in the framework of (and funded through) the CDOP-2 phase of the EUMETSAT Ocean and Sea Ice SAF. Current Status 1.Will use new processing chain and algorithms making use of ECMWF outputs) 2.Reprocessing is scheduled for 2013/2014 3.SST files will be produced a hourly GDS v2.0 L3C, collating data from 15 minutes slots in 1 file Looking Forward 1.Data will be ready for delivery after 2014 Project Summary: GHRSST-12 Reanalysis Breakout Session Hervé Roquet, Meteo France

27. 27 Level 4 Activities

28. 28 Project Goals 1.Provide consistent long L4 SST daily timeseries for regional climate modelling, and climatologies to be used as background (first guess) field for operational real-time L4 processing. (MERSEA-IP, PRIMI, MyOCEAN) 2.Assess the system variability and trends via statistical analyses (SESAME-IP) 3.Build robust climatologies for comparison and validation of in-situ climatologies (SeaDataNet) http://gos.ifa.rm.cnr.it/index.php?id =407 Current Status 1.Version 1 Available: Daily 1985-2005, and climatologies 2.Available via FTP, HTTP 3.Climatologies are used as first guess for the L4 operational production within MyOCEAN SST-Tematic Assembly Center Looking Forward 1.Working on ultra high resolution product now 2.Available late 2010 Mediterranean and Black Sea re-analyses Bruno Buongiorno Nardelli, GOS/CNR Project ON HOLD Project Summary: GHRSST-12 Reanalysis Breakout Session

29. 29 OSI SAF Sea Ice Concentration Project Goals - reprocess all available PMW data (SMMR, SSM/I) - provide daily sea ice concentration back to 1978 Current Status version 1.0 was released spring 2010 (1978-2007) version 1.1 in progress (1978-2009/2011), with improved uncertainties Looking Forward - continuous updates from end 2012 - more information on http://osisaf.met.no Project Summary: GHRSST-12 Reanalysis Breakout Session Steinar Eastwood, met.no

30. 30 HadISST2 Project Goals 1.A global, gridded, reconstructed data set of in situ / AVHRR / ATSR SST and sea ice measurements since 1850 with reduced biases and quantified uncertainties, including an assessment of their covariance structure Relevant Graphic Current Status 1.Prototype to be available autumn 2011 Looking Forward 1.Will incorporate ATSR Reanalysis for Climate (ARC) retrievals in final version, due autumn 2012. Project Summary: GHRSST-12 Reanalysis Breakout Session Nick Rayner, UKMO

31. 31 MUR Phase I Project Goals Multi-scale analysis at ~1 km resolution. Phase I covers the Aqua satellite era (2002 to present). Phase II extension in time using HRPT data. Current Status Version 2 covers from late 2008 to near-present (May 2011); production is on-going. GLOB / JPL / MUR available at PO.DAAC Looking Forward Version 3 is under plan to optimize analysis scales. Near real-time production. Incorporation of a motion- compensation algorithm. Project Summary: GHRSST-12 Reanalysis Breakout Session Mike Chin, JPL

32. 32 NOAA NESDIS Blended SST Analysis MUR Relevance to NASA Goals Looking Forward Submitted a proposal to reprocess Project Goals Improve the analysis by implementing changes that will improve the accuracy of the analysis Reprocess 0.25 Degree POES-GOES Blended Analysis Single vs multi- resolution analysis from the same data using the same finest-level analysis scale. Date: Jan 30, 2008 Current Status Improved error catagerization Inclusion of microwave SST AATSR SST for reference field Project Summary: GHRSST-12 Reanalysis Breakout Session

33. 33 DMI L4 Reanalysis, with Greenland Research Climate Centre Project Goals A L4 SST reanalysis in 4 km 1985 to present, based upon Pathfinder and ARC data Cover Arctic and focus upon Greenland waters around Nuuk. Use the OSI SAF reanalysis ice mask. Including ICOADS2.5 + in situ observations from Greenland Area Current Status Reanalysis planned to be carried out Summer-Fall 2011 Preliminary release, December 2011. Looking Forward Trends and anomalies will be derived from the dataset during 2011 and compared to in situ time series in Greenland. Jacob Hoyer, DMI Project Summary: GHRSST-12 Reanalysis Breakout Session

34. 34 ESA SST Climate Change Initiative Project Goals 1.Generate an SST CDR for 1991 to 2010 using ATSR + AVHRR 2.CDR features: independent of in situ; homogenized and corrected for diurnal cycle; 0.05 deg resolution; 0.05 K / dec stability goal; GHRSST formats; skin and drifter-depth estimates; uncertainties modelled 3.Prototype, demonstrate & specify a system for ongoing CDR in future Relevant Graphic Current Status 1.User requirements collected 2.Product specifications available for RAN-TAG / GHRSST comment 3.Open / competitive algorithm selection process underway, based on multi-sensor matchup data set 4.Input data gathered for processing Looking Forward 1.Algorithm selection process open until October 2011. 2.Prototyping in 2012. 3.Data distribution in 2013. Project Summary: GHRSST-12 Reanalysis Breakout Session Chris Merchant

35. 35 OSTIA Reanalysis OSTIA reanalysis produces global, daily SST and sea-ice reanalysis at high resolution (1/20°). Reanalysis runs from 1985-2007. AVHRR Pathfinder and ATSR-1 data are bias corrected. The in-situ ICOADS and ATSR-2/AATSR data used as reference data set. OSI-SAF sea-ice concentration reprocessing used. SST trends 1985-2007 (°K/yr). -0.25 0.25 OSTIA reanalysis completed. Data freely available through MyOcean project http://www.myocean.eu.org/ OSTIA reanalysis been assessed and validated in Technical Report (in review). Paper will be written on the OSTIA reanalysis. Next version produced as part of the ESA SST CCI project. Satellite data will be produced within the project. In-situ data not assimilated but used to validate the reanalysis. Project Summary: GHRSST-12 Reanalysis Breakout Session Jonah Roberts-Jones, Met Office

36. 36 Nighttime-only OISST Project Goals 1.To make separate optimally interpolated day and night SST fields that have reduced diurnal variability than a daily OI Relevant Graphic Current Status 1.Preliminary comparisons with 2004- 2006 Pathfinder v.5.1 indicate daytime-only OI agrees better with daytime AVHRR than daily OI 2.Daily OISST similar to nighttime only OISST 3.Seasonality in difference with AVHRR was not eliminated Looking Forward 1.Use new Pathfinder SST 2.Examine bias correction Project Summary: GHRSST-12 Reanalysis Breakout Session PFDay minus Daytime only OI Viva Banzon, NOAA NCDC

37. 37 Project Goals 1.Provide consistent time series of global daily SST analysis during the satellite era 2.Used for ocean reanalysis with a ocean data assimilation system for the western North Pacific (MOVE/MRI.COM-WNP) New Climatologies of SST Current Status 1.SSTs for 1985-2004 was analyzed using Pathfinder V5 2.Operational recalculation (delayed about 5 months) since 2005 using AMSR-E (JAXA) and AVHRR (GAC) 3.Update Climatologies of SST for 1981-2010 using MGDSST and COBE-SST Looking Forward 1.Reanalysis for 1981-1984 using Pathfinder V5.2 (under processing) and revise climatologies 2.Compare with another of JMA`s SST analyses (COBE-SST), which has long term record (> 100yrs) and uses only in-situ observation MGDSST Reanalysis Shiro Ishizaki, JMA differences in climatology for July (new-old) new climatology for July (1981- 2010) Project Summary: GHRSST-12 Reanalysis Breakout Session

38. 38 LDEO [Kaplan et al., 1998, 2003] SST Project Goals To provide a statistically homogenous analysis of the full historical data set of in situ observations, with verified uncertainty estimates, using realistic covariances of the “true” field and data error, with the minimum of other assumptions. To support studies of climate variability and paleoreconstruction verification; serve as a benchmark for other century-scale analyses. Current Status The original 5 o x5 o analysis version of for in situ data only (1856--) extended to present using monthly NCEP OI v.2 is updated operationally by IRI Data Library; part of the GCOS intercompa-rison; finishing GCOS test experim. Recent methodological developments: adding smaller scales to the reduced space solution, Bayesian ensemble representation of uncertainty (w/ A.Karspeck & S.Sain, NCAR). Looking Forward Using SST variability estimates from satellite data to model in situ data error and use it in a reduced space analysis of historical data; adding middle scales of SST variability to the global large scale solution at 1 o x1 o resolution; practical representation of the uncertainty by an ensemble of realizations; making this ensemble available on the web (collab w/A.Karspeck & S.Sain, NCAR). http://iridl.ldeo.columbia.edu/expert/SOURCES/.KAPLAN/.EXTENDED/.v2/.ssta/ Alexey Kaplan, LDEO

39. 39 Data Tools/Intercomparison Activities

40. 40 GCOS SST Intercomparison Framework Looking Forward 1.Receive rest of common-input analyses for pilot study 2.Update products with new versions (e.g. ERSSTv3b) and new data (e.g. PFV5.2 through 2010) 3.Summarize differences as a set of strengths and weaknesses for each product to provide guidance to users of SST analysis products Current Status 1.11 unique products in the framework, as well as standard diagnostics and browse graphics 2.All GCOS data cubes available via NODC Live Access Server 3.As part of pilot study to determine the origins of observed differences, common input data and two historical analyses (ERSST and COBE) based on that input received Project Goals 1.To understand differences between global and near-global SST analyses with the aim of producing better, long-term SST climate data records 2.To use this understanding to link the modern satellite-based records with the historical, primarily in situ-based records http://ghrsst.nodc.noaa.gov/intercomp.html Tess Brandon, NOAA NODC Project Summary: GHRSST-12 Reanalysis Breakout Session

41. 41 2011 GHRSST Data Availability Pierre LeBorgne Ken Casey and Bob Evans Anne O'Carroll Peter Cornillon, Helen Beggs Andy Harris and Eileen Maturi Jorge Vazquez Andy Harris and Eileen Maturi Chelle Gentemann Chris Metchant Pierre LeBorgne

42. 42 3. Establishing A Data Processing Framework for the SST ECV

43. 43 We already have: GDS2 and the GHRSST R/GTS FrameworkGDS2 and the GHRSST R/GTS Framework Decent international coordination (RAN- TAG, GCOS SST and Sea Ice WG, etc)Decent international coordination (RAN- TAG, GCOS SST and Sea Ice WG, etc) A Conceptual FrameworkA Conceptual Framework

44. 44 SST ECV Conceptual Framework 44 L4 L3 L2 L1 L0 CoarseLess Coarse FineFiner Fundamental CDRs Thematic CDRs SST_depth SST_fnd SST_subskin SST_skin

45. 45 But we need more than just a Conceptual Framework…

46. 46 ECV Data Processing Framework Define and document a clear framework for data providers who wish their datasets to be considered and evaluated as SST ECVs –Go beyond GDS2 format and the GDAC/LTSRF data management system Once defined and documented, build and support the framework

47. 47 What should be in the ECV DPF?

48. 48 GCOS Climate Monitoring Principles http://www.wmo.int/pages/prog/gcos/documents/GCOS_Climate_Monitoring_Principl es.pdf http://www.wmo.int/pages/prog/gcos/documents/GCOS_Climate_Monitoring_Principl es.pdf 1.The impact of new systems or changes to existing systems should be assessed prior to implementation. 2.A suitable period of overlap for new and old observing systems is required. 3.The details and history of local conditions, instruments, operating procedures, data processing algorithms and other factors pertinent to interpreting data (i.e., metadata) should be documented and treated with the same care as the data themselves. 4.The quality and homogeneity of data should be regularly assessed as a part of routine operations 5.Consideration of the needs for environmental and climate-monitoring products an assessments, such as IPCC assessments, should be integrated into national, regional an global observing priorities 6.Operation of historically-uninterrupted stations and observing systems should be maintained. 7.High priority for additional observations should be focused on data-poor regions, poorly observed parameters, regions sensitive to change, and key measurements with inadequate temporal resolution. 8.Long-term requirements, including appropriate sampling frequencies, should be specified to network designers, operators and instrument engineers at the outset of system design and implementation. 9.The conversion of research observing systems to long-term operations in a carefully-planned manner should be promoted. 10.Data management systems that facilitate access, use and interpretation of data and products should be included as essential elements of climate monitoring systems.

49. 49 (GCOS Satellite Supplement) 11.Constant sampling within the diurnal cycle (minimizing the effects of orbital decay and orbit drift) should be maintained. 12.A suitable period of overlap for new and old satellite systems should be ensured for a period 13.Continuity of satellite measurements (i.e. elimination of gaps in the long-term record) through appropriate launch and orbital strategies should be ensured. 14.Rigorous pre-launch instrument characterization and calibration, including radiance confirmation against an international radiance scale provided by a national metrology institute, should be ensured. 15.On-board calibration adequate for climate system observations should be ensured and associated instrument characteristics monitored. 16.Operational production of priority climate products should be sustained and peer-reviewed new products should be introduced as appropriate. 17.Data systems needed to facilitate user access to climate products, metadata and raw data including key data for delayed-mode analysis, should be established and maintained. 18.Use of functioning baseline instruments that meet the calibration and stability requirements stated above should be maintained for as long as possible, even when these exist on decommissioned 19.Complementary in situ baseline observations for satellite measurements should be maintained through appropriate activities and cooperation. 20.Random errors and time-dependent biases in satellite observations and derived products should be identified. GCOS Climate Monitoring Principles

50. 50 NOAA CDRP Maturity Model Perros-Guirec, June 2008

51. 51 ESA CCI Climate Ops Model Perros-Guirec, June 2008

52. 52 Requirements from SST_cci URD Global coverage Regular latitude/longitude grid Records longer than 30 years Routinely updated Bias no more than 0.1K, demonstrated over ideally 100km regions (but see slide 52) Precision no worse than 0.1K, demonstrated over ideally 100km sub-regions (but see slide 53) Drift no worse than 0.1K/decade, demonstrated over ideally 100km sub-regions (but see slide 54) Drift over the annual cycle of no more than 0.1K/decade, demonstrated over ideally 100km sub- regions Full characterisation of uncertainties and their correlation structure Uncertainties provided with the SST, split into random and correlated components Uncertainties to be verified against independent observations Provision of information on windspeed and the location of clouds and sea ice Address the needs of users with least developed computing infrastructure Data provided free at point of use Data provided to download through a variety of means Data provided in NetCDF Full publication of all steps in development and peer reviewed publication A statement saying point by point which GCOS guidelines have been followed should be published. Users should have easy access to information, documents, products and contacts through a high quality website. Verification against independent data and comparison to other equivalent products. Metadata describing data sources. Defining the SST ECV DPF: GHRSST-12 Reanalysis Breakout Session

53. 53 Perros-Guirec, June 2008 RAN-TAG/GCOS Requirements https://www.ghrsst.org/ghrsst-science/science-team- groups/ran-tag/latest-climate-record-requirements/ Accuracy, delivery timescale, uncertainty estimates, spatial and temporal characteristics, temporal stability, SST product types, data format, metadata, archive and distribution, and target accuracy

54. 54 Acceptable levels of bias http://www.esa-sst-cci.org/ From technical note to Product Specification Document

55. 55 Acceptable levels of precision http://www.esa-sst-cci.org/ From technical note to Product Specification Document

56. 56 Acceptable levels of drift http://www.esa-sst-cci.org/ From technical note to Product Specification Document

57. 57 Possible Elements of a DPF Data in GDS2 format Data submitted to GDAC/LTSRF Data in GHRSST/GCOS Intercomparison Facility Data assessed using community-established metrics –Should measure product against GCOS/ESA CCI/RAN-TAG requirements –Metrics should be accessible –Metrics such as o-b bias and error –What else? Perros-Guirec, June 2008

58. 58 Possible Elements of a DPF Production of Reanalysis GMPE to inter-compare L4 products (produced as part of ESA SST CCI). Documentation and description of algorithms used and validation/assessment metrics conducted Version control of accessible, well-documented code Full metadata including lineage/provenance Reprocessing update alert system Perros-Guirec, June 2008

59. 59 Data assessed against independent, QC’d in situ SST (e.g. Argo, drifting buoys, depth SST from ships of reliable accuracy like IMOS ships) Data assessed against an established, traceable, reliable network of in situ radiometers (e.g. ISAR, M- AERI) Perros-Guirec, June 2008

60. 60 Questions/Issues Data in NetCDF using GHRSST variable names and definitions (ie, is there a slight step back from fully compliance with GDS2.0 that would still work and maximize inclusivity?) [KC: Suggest alternate approach, via updates to GDS] Briefing note (few pages) giving key characteristics of data, known strengths and weaknesses … : we should plan a ToC for this? [Being done in DAS-TAG] ECVs – in GMPE, or assessed against GMPE only? ECVs - in SQUAM? In HR-DDS? If the metrics are in onerous, it is difficult to insist that they be computed. Defining the SST ECV DPF: GHRSST-12 Reanalysis Breakout Session Possible Elements of a DPF

61. 61 4. Priority Actions for the Coming Year

62. 62 Prioritized List of Future Actions Intercomparisons Define the ECV DPF Determine how GCOS SST Intercomparison Facility should be involved in the SST ECV DPF Complete GCOS SST Intercomparison pilot study –Generate SST analyses using common QC’ed input UKMO data –Incorporate analyses into intercomp facility and evaluate results Explore NIST traceability for SST CDRs Assess product accuracy using consistent independent SST data set such as Argo Establish in situ radiometer network

63. 63 Prioritized List of Future Actions Products Pathfinder V5.2 (GAC L2P, L3U, and L3C) Pathfinder V6 progress (GAC and LAC/HRPT) (A)ATSR, MODIS, and GOES reprocessed data in GDS2 An accurate, long-term, daily foundation SST analysis from IR and microwave SST and best available processing methods Reanalysis data sources in common format. (Being done for ESA SST CCI, would be useful to include in-situ obs.) Identifying a subset of reference data for reanalysis period. Reanalysis being done for ESA SST CCI for 1991 onwards, is it needed back to 1981? AVHRR GAC/LAC L1b consolidated, calibrated, geolocated archive – e.g., ongoing NOAA CDR Program, SCOPE, R2HA2 Assess and provide feedback on ATSR Reprocessing for Climate (ARC) v1.0/v1.1 SST release Perros-Guirec, June 2008

64. 64 Prioritized List of Future Actions Algorithms/Science/Uncertainties Improving L4 algorithms and analysis error estimation in L4 products. Bias and uncertainty estimates in reanalysis in-situ data for each ob. (May be carried out as ICOADS IVAD but several different estimates exist?) Improve bias correction of L4 products (particularly during Mount Pinatubo period). Fix the calibration of AVHRR and AMSR-E to remove latitudinal and other biases before reprocessing of data sets Quantify uncertainties and their covariance structure through bottom-up approach, i.e. understanding the uncertainties inherent in each step of data production Determine how to assess and compare stability of satellite SST CDRs Perros-Guirec, June 2008

65. 65 Prioritized List of Future Actions User Tools Provide guidance/recommendations to non-expert SST users in choosing from an overwhelming variety of products – GHRSST website, GDAC, LTSRF Provide an interactive look-up table for GDAC/LTSRF L2P, L3, L4 and L4S products to help users to choose the most appropriate SST product for their application from a range of criteria such as 1.Date range 2.Temporal resolution 3.Effective spatial resolution (not just grid resolution) 4.Depth of SST (skin, subskin, blend, foundation) 5.Level 2 (single swath), level 3 (composite), level 4 (analysis) or ensemble of level 4 analyses Perros-Guirec, June 2008

66. 66 5. New RAN-TAG Chair Dr. Chris Merchant University of Edinburgh