1 / 29

Global Drifter Program (GDP)

Global Drifter Program (GDP). Drifting buoy measurements of Sea Surface Temperature, Mixed Layer Currents, Atmospheric Pressure and Winds http://www.aoml.noaa.gov/phod/dac/gdp.html. Luca Centurioni, SIO/CIMEC & CASPO Rick Lumpkin, NOAA/AOML. OUTLINE

matthewwalden
Télécharger la présentation

Global Drifter Program (GDP)

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Global Drifter Program (GDP) Drifting buoy measurements of Sea Surface Temperature, Mixed Layer Currents, Atmospheric Pressure and Winds http://www.aoml.noaa.gov/phod/dac/gdp.html Luca Centurioni, SIO/CIMEC & CASPO Rick Lumpkin, NOAA/AOML • OUTLINE • Status of the GDP array and program’s highlights • Evaluation of the impact of SLP data from drifters on Numerical Weather Prediction • Drogue loss re-evaluation and proposed actions NOAA Climate Observation Division, 8th Annual PI Meeting ,24-27 June 2012, Silver Spring, MD

  2. GDP purpose GDP: the principal component of the Global Surface Drifting Buoy Array, a branch of NOAA’s Global Ocean Observing System (GOOS) and Global Climate Observing System (GCOS) and a scientific project of the Data Buoy Cooperation Panel (DBCP). Objectives: Maintain a global 5ºx5º array of ~1250 satellite-tracked Lagrangian surface drifting buoys to meet the need for an accurate and globally dense set of in-situ observations: mixed layer currents, SST, atmospheric pressure, winds, and salinity. Provide data processing system for scientific use of these data. These data support short-term (seasonal-to-interannual) climate predictions as well as climate research and monitoring.

  3. FUNDING: NOAA’s Climate Program Office (additional instrument development at Scripps supported by ONR) Organization of the Global Drifter Program Manufacturers in private industry, who build the drifters according to closely monitored specifications Scripps (La Jolla, CA) Luca Centurioni AOML (Miami, FL) Rick Lumpkin Industry supervision, purchase/build ~900 drifters/year Drifter Operations Center (DOC) Drifter Data Assembly Center (DAC) Develop enhanced velocity data sets Upgrade the technology & new sensors

  4. Current status of the global array ~988 GDP drifters (goal: 1250 annual average) Goal: 1250 total, 1000 by NOAA, 250 by partners (20%). 867 NOAA (87% target), 121 partners (48% target)

  5. Science applications • 65 peer reviewed papers using GDP data published 2011-present (~800 total) • Technical developments (overall goal is to homogenize the SVP design) • Ruggedized tether attachment for strength and water infiltration implemented across fleet; • Recommendation for high quality batteries issued to manufacturers; • Recommendation for more accurate STT (0.05°C) issued to manufacturers; • Recommendation for ruggedized drogue design issued to manufacturers; • SIO completed SVP and SVPB drifter design and started production; • New tether material (synthetic rope) is currently under evaluation (20 SIO units); • Deployments • 24 PG SVPB's in the Drake Passage -Christian Reiss's annual cruise • 20 Clearwater SVPB's near Falkland Islands -Andrew Thompson • 10 Metocean SVPB's from PX8-DBCP Iridium Upgrades • 12 PG SVPB's in E. Indian-Indonesian Colleagues • 20 Clearwater SVP's during Latmix II - Pascale Lelong • 20 Clearwater SVPB's in S. Indian Ocean - IsabellAnsorge • 20 SIO SVP's off coast of California – Sam McClatchie • 16 SVP's in conjunction with Earth Day - 6 US Cities • 10 DBI SVP's from PX8 • 10 Clearwater SVP's in Yucatan Strait - EstrellaMalca FY2011-2012 Program Highlights

  6. GDP array decline since mid 2010. Below1250 threshold in early 2011

  7. Recent deployments • Total for Oct 1 2011—June 17 2012: 783 drifters • 2009:1035 (DOC 944) • 2010:1279 (DOC 1173) • 2011: 1218 (DOC 976) • 2012: 332 (as of May 18, DOC 234)

  8. Evaluation of the Global Drifter Array: troubleshooting the anomalous death rate • Accelerated array decline appeared in fall ‘11; • Already known (‘10&’11) and addressed cause was defective (leaking) battery packs (2 companies); • Second cause, investigated (and addressed) in winter ‘11/’12 was “unconventional” design changes (1 company); • Third cause pertains to higher energy consumption of new Argos 3 transmitter (PMT) when used in Argos 2 mode.

  9. How to fix a fleet of power hungry devices.Options: • Increase size of battery pack: partial fix, applicable only for large hulls; • Switch to Argos 3: ok, but only one Argos 3 satellite is active. Potential issue if A3 satellite fails; • Change satcom mode to Iridium. • Efforts are underway to implement 1&2 • Array has been inching up in the past month

  10. IMPACT OF SLP DATA FROM DRIFTERS ON NWP • Workshop held on May 21, 2012 in Sedona, AZ: • Summary of conclusions: • The impact of drifters (i.e. from fraction of beneficial observations, impact per observations and many others indicators) is very high for different assimilation systems (3D-Var, 4D-Var) and different global metrics; • The impact is large especially for open ocean, high latitudes buoys (i.e. drifters) and for explosive cyclogenesis; • There is little redundancy in the SLP observing system (high % of beneficial observations); • Drifter SLP is very important to anchor the SLP pressure field and is more accurate than GPSRO; • Need to run routine impact studies with drifters’ SLP separated from other marine network components and for latitude bands; • Introduce also alternate metrics, although not optimal in a global sense, such as surface winds, of more direct impact to society; • Summarize workshop discussion in a BAMS paper; • Evaluate need to run more focused OSE (drifters only).

  11. Definition of Observation Impact following Langland and Baker (2004); extended for nonlinear analysis schemes by Trémolet (2008) Forecast Verifying Analysis Analysis Background Observation Impact: δe < 0 …the observation(s) improve the forecast Courtesy of Ron Gelaro, NASA-GFSC

  12. Impact Impact / Obs. number Forecast impact experimentfromDec. 2010 to Jan. 2011 (4D VAR) BUOY BUOY Courtesy of Jean-Francois Mahfouf, MeteoFrance

  13. Gridded Observation Total Impact January 2012 Raob u,v,T,q Buoy ps NWS Winter Storms Reconnaissance Targeting u,v only, mostly lower-trop (700-900 hPa) Pibal u,v Drop u,v,T,q Courtesy of Ron Gelaro, NASA

  14. Gridded Observation Total Impact July 2011 Raob u,v,T,q Buoy ps Pibal u,v Drop u,v,T,q Courtesy of Ron Gelaro, NASA

  15. Results: SP-Denial versus Control Control better Relative error diff Denial better SLP OSE RUN AT ECMWF: COURTESY OF CARLA CARDINALI WMO DBCP Workshop Sedona 2012

  16. Mixed layer current measurementsThe drogue is essential to ensure the Lagrangian behavior of the drifters Black Bullets: location of drogued (622) and un-drogued (366) drifters . Open pink: drifters without drogues, still measuring SST & SLP (if fitted). ~63% of the array has drogues attached. In line with historical pre-2005 assessment, but still needs improvement.

  17. Drogue loss How do we know if the drogue is attached? Submergence (orig. method) tether strain (new method)

  18. Evidence of problems Left: time-mean zonal currents from drifters and two independent analyses (Grodsky et al., 2011). Geostrophic component can be calculated from AVISO and CLS mean dynamic height (Rio et al.., 2011).

  19. How can we fix the problem? Short-term fix recommended by Grodsky et al. (2011): use only first 90 days of drifters for Jan 2004—December 2008. Eliminates 90% of data during this period! Rio (2012, submitted to J. Tech.): A (latitude, month) chosen to minimize residuals for all “drogue on” drifters <90d old. Then, for any drifter, With drogue attached,~0 (by construction). Drogue lost: = 0.015—0.02.

  20. Fixing the problem Drogue off originally diagnosed after 1080 days Drogue off actually after ~560 days

  21. ID 6140 Automatic (first-pass) reanalysis Follows methodology inspired by Marie-Hélène Rio. Least-squares fit of step function to determine where a increases. Results: 8.7k of 13.6k drifters (64%) : < 30d difference. a method failed to identify 2055 “known” drogue loses. 2848 drifters (21%) have drogue off date earlier by >30 days. 18% reduction in “drogue on” data for period Oct 1992—Dec 2010. “Signal” in some submergence records that can be reinterpreted.

  22. original Comparison of data: before/after automatic reanalysis reanalysis reanalysis original original reanalysis

  23. Rio et. al (2012) geostrophic model or similar formulations (Niiler et al. 2003, Centurioni et al. 2008, Maximenko et al. 2009) only perform when altimetry-derived geostrophic current anomalies correlate well with drifter data

  24. Example of “hidden signal” in submergence Days after deployment

  25. Why drifters loose the drogue (besides fish bites)? Note several kinks. Breaking points + rust SIO drifters recovered in Tuscany (Italy) and... SIO is evaluating 20 drifters with new tether material (synthetic rope) Damage due to ~1 week on cliff …off Tigzirt (Algeria) SIO specified drogue (enhanced stitching, tacking, doubled folds)

  26. Summary • GDP array has been declining since mid 2010 due to technical issues, now addressed, and is on the way to recovery; • Impact of SLP data from GDP on NWP evaluated and shown to be very significant. No obvious redundancy is apparent in the array. More investigations needed (impact time series for drifters only, new OSE, pub) • Drogue status re-evaluated with new indirect techniques, dataset re-processed (automatic & manual) and available in beta mode through AOML. Pilot array with 20 SIO built synthetic rope tethers will be deployed in August 2012 to improve drogue life

  27. 2012-2013: goals and plans • Deploy 1000 Drifters in the period between October 2011 and September 2012. • RESTORE 1250 drifters at a nominal resolution of 5° x 5°. • Continue to update quality-controlled interpolated database. • Evaluate performance of various manufacturers’ drifters. • Finalize impact studies of drifters’ SLP data impact. • Finalize selection of tether material. • Continue to evaluate array evolution, drogue presence and drogue lifetime. • Explore new SSS sensors, transition QC procedures for SSS drifter data to operation. • Develop new products, primarily velocity dataset with drogue loss reevaluation.

  28. Our appreciation to the following partners for their contributions to GDP activities Thanks Institute of Hydrological and Oceanic Services, National Sun Yet-Sen University (Taiwan) Centro de Investigacion Cientifica y de Educacion Superior de Ensenada (Mexico) Korean Oceanographic Research and Dvelopment Institute, National Oceanographic Research Institute; Ministry of Maritime Affairs and Fisheries (Korea) Instituto del Mar del Peru Tristan da Cunha Administration, Tristan Island United Kingdom Met Office Fisheries Department of Falkland Islands Environment Canada University of Cape Town; South African Weather Service (South Africa) Scripps Institution of Oceanography, Woods Hole Oceanographic Institution, Oregon State University, Marine Resources Research Institute, NOAA/Pacific Marine Environmental Laboratory, NOAA/National Data Buoy Center (United States of America) US Office of Naval Research (ONR) United States Air Force US Naval Oceanographic Office United States Coast Guard Raytheon Polar Services NOAA’s Voluntary Observation Ships, Ships of Opportunity, and National Marine Fisheries Service programs Argo program International Ice Patrol Institut de Recherche pour le Développement; Météo-France (France) ENSSMAL and Protection Civile (Algeria) Leibniz-Institut für Meereswissenschaften an der Universität Kiel (Germany) New Zealand Met. Service Australian Bureau of Meteorology Fundação Universidade Federal do Rio Grande; Instituto Nacional de Metereologia; Centro de Hydrografia de Marinha; INPE (Nacional Space Institute); Brazilian Navy; Brazilian Naval Directorate of Hydrography and Navigation (Brazil) Fisheries Research Institute; Servicio de Hidrografía Naval (Argentina) Instituto Canario de Ciencias Marinas; Universidad de Las Palmas de Gran Canaria (Spain) Instituto Nazionale di Oceanografia e di Geofisica Sperimentale, LAMMA (Italy) National Institute of Oceanography; National Institute of Ocean Technology (India)

  29. Note the median age ranges between 250-330 days since 2008

More Related