Pg 1 A Loosely Coupled Ocean-Atmosphere Ensemble Assimilation System. Tim Hoar, Nancy Collins, Kevin Raeder, Jeffrey Anderson, NCAR Institute for Math.

Slides:



Advertisements
Similar presentations
Parameter Sensitivity of a Coupled Climate Model
Advertisements

Integrated Earth System Analysis Workshop Report Held November 2010, Baltimore MD Details at 2PM – POS Panel.
Experiments with Monthly Satellite Ocean Color Fields in a NCEP Operational Ocean Forecast System PI: Eric Bayler, NESDIS/STAR Co-I: David Behringer, NWS/NCEP/EMC/GCWMB.
INPE Activities on Seasonal Climate Predictions Paulo Nobre INPE-CCST-CPTEC WGSIP-12, Miami, January 2009.
Verification of NCEP SFM seasonal climate prediction during Jae-Kyung E. Schemm Climate Prediction Center NCEP/NWS/NOAA.
1 Evaluation of two global HYCOM 1/12º hindcasts in the Mediterranean Sea Cedric Sommen 1 In collaboration with Alexandra Bozec 2 and Eric Chassignet 2.
Initialization Issues of Coupled Ocean-atmosphere Prediction System Climate and Environment System Research Center Seoul National University, Korea In-Sik.
Jon Robson (Uni. Reading) Rowan Sutton (Uni. Reading) and Doug Smith (UK Met Office) Analysis of a decadal prediction system:
Geophysical Fluid Dynamics Laboratory Review June 30 - July 2, 2009 Geophysical Fluid Dynamics Laboratory Review June 30 - July 2, 2009.
Instituting Reforecasting at NCEP/EMC Tom Hamill (ESRL) Yuejian Zhu (EMC) Tom Workoff (WPC) Kathryn Gilbert (MDL) Mike Charles (CPC) Hank Herr (OHD) Trevor.
Enhanced seasonal forecast skill following SSWs DynVar/SNAP Workshop, Reading, UK, April 2013 Michael Sigmond (CCCma) John Scinocca, Slava Kharin.
Empirical Localization of Observation Impact in Ensemble Filters Jeff Anderson IMAGe/DAReS Thanks to Lili Lei, Tim Hoar, Kevin Raeder, Nancy Collins, Glen.
THE BEST ANALYZED AIR- SEA FLUXES FOR SEASONAL FORECASTING 2.12 Glenn H. White, W. Wang, S. Saha, D. Behringer, S. Nadiga and H.-L. Pan Global Climate.
The NCEP operational Climate Forecast System : configuration, products, and plan for the future Hua-Lu Pan Environmental Modeling Center NCEP.
A Regression Model for Ensemble Forecasts David Unger Climate Prediction Center.
© 2012 National Ecological Observatory Network, Inc. ALL RIGHTS RESERVED. THE DATA ASSIMILATION RESEARCH TESTBED (DART) FOR ECOLOGICAL FORECASTING Andy.
Possible Strategic Capability Projects High resolution coupled simulation: (ASD) – 1/8 o CAM, 1/10 o POP, 20 years (initially) – SCIDAC proposal: Small,
Ensemble Data Assimilation and Uncertainty Quantification Jeffrey Anderson, Alicia Karspeck, Tim Hoar, Nancy Collins, Kevin Raeder, Steve Yeager National.
Open Source Ensemble Kalman Filtering: the Data Assimilation Research Testbed - DART Tim Hoar, Jeffrey Anderson, Nancy Collins, Kevin Raeder, Hui Liu,
Collaborative Research: Toward reanalysis of the Arctic Climate System—sea ice and ocean reconstruction with data assimilation Synthesis of Arctic System.
The Centre for Australian Weather and Climate Research A partnership between CSIRO and the Bureau of Meteorology Summary/Future Re-anal.
Ocean-atmosphere interaction in the Atlantic storm track response to climate change Tim Woollings With Jonathan Gregory, Joaquim Pinto, Mark Reyers and.
1 Geophysical Fluid Dynamics Laboratory Review June 30 - July 2, 2009.
Inter-comparison and Validation Task Team Breakout discussion.
NACLIM CT1/CT3 1 st CT workshop April 2013 Hamburg (DE) Johann Jungclaus.
Tim Hoar : National Center for Atmospheric Research with a whole lot of help from: Jeff Anderson, Nancy Collins, Kevin Raeder, Bill Sacks : NCAR Yongfei.
Reanalysis: When observations meet models
Potential impact of HF radar and gliders on ocean forecast system Peter Oke June 2009 CSIRO Marine and Atmospheric Research.
Development of an EnKF to estimate CO 2 fluxes from realistic distributions of X CO2 Liang Feng, Paul Palmer
Ben Kirtman University of Miami-RSMAS Disentangling the Link Between Weather and Climate.
The European Heat Wave of 2003: A Modeling Study Using the NSIPP-1 AGCM. Global Modeling and Assimilation Office, NASA/GSFC Philip Pegion (1), Siegfried.
AIRS Radiance and Geophysical Products: Methodology and Validation Mitch Goldberg, Larry McMillin NOAA/NESDIS Walter Wolf, Lihang Zhou, Yanni Qu and M.
Jim Hurrell OceanObs September 2009 OceanObs09: Ocean Information for Society Day III: Delivering Services to Society Session B: Forecasting Decadal-to-Centennial.
OCO 10/27/10 GFDL Activities in Decadal Intialization and Prediction A. Rosati, S. Zhang, T. Delworth, Y. Chang, R. Gudgel Presented by G. Vecchi 1. Coupled.
Tim Hoar: National Center for Atmospheric Research with a whole lot of help from: Bill Sacks, Mariana Vertenstein, Tony Craig, Jim Edwards: NCAR Andrew.
Statistical Post Processing - Using Reforecast to Improve GEFS Forecast Yuejian Zhu Hong Guan and Bo Cui ECM/NCEP/NWS Dec. 3 rd 2013 Acknowledgements:
Multi-Model Ensembles for Climate Attribution Arun Kumar Climate Prediction Center NCEP/NOAA Acknowledgements: Bhaskar Jha; Marty Hoerling; Ming Ji & OGP;
An Examination Of Interesting Properties Regarding A Physics Ensemble 2012 WRF Users’ Workshop Nick P. Bassill June 28 th, 2012.
Preliminary results from assimilation of GPS radio occultation data in WRF using an ensemble filter H. Liu, J. Anderson, B. Kuo, C. Snyder, A. Caya IMAGe.
One-year re-forecast ensembles with CCSM3.0 using initial states for 1 January and 1 July in Model: CCSM3 is a coupled climate model with state-of-the-art.
Ocean Climate Simulations with Uncoupled HYCOM and Fully Coupled CCSM3/HYCOM Jianjun Yin and Eric Chassignet Center for Ocean-Atmospheric Prediction Studies.
Alan F. Hamlet Andy Wood Dennis P. Lettenmaier JISAO Center for Science in the Earth System Climate Impacts Group and the Department.
PreSAC Progress on NEMOVAR. Overview of NEMOVAR status First NEMOVAR experiments Use of NEMOVAR analyses to initialize ocean only forecasts Missing.
Impact of TAO observations on Impact of TAO observations on Operational Analysis for Tropical Pacific Yan Xue Climate Prediction Center NCEP Ocean Climate.
Pg 1 SAMSI UQ Workshop; Sep Ensemble Data Assimilation and Uncertainty Quantification Jeff Anderson National Center for Atmospheric Research.
Future Directions in Ensemble DA for Hurricane Prediction Applications Jeff Anderson: NCAR Ryan Torn: SUNY Albany Thanks to Chris Snyder, Pavel Sakov The.
Understanding Forecast Skill What is the Overall Limit of Predictability? What Limits Predictability? –Uncertainty in Initial Conditions: Chaos within.
CESM-CAM5 Large Ensemble Update May 13, list for all future updates -
ESSL Holland, CCSM Workshop 0606 Predicting the Earth System Across Scales: Both Ways Summary:Rationale Approach and Current Focus Improved Simulation.
Pg 1 ESPC DA Workshop; Sept The Data Assimilation Research Testbed: A Community Ensemble DA Facility Jeffrey Anderson, Nancy Collins, Tim Hoar, Hui.
Coupled Initialization Experiments in the COLA Anomaly Coupled Model
Jeffrey Anderson, NCAR Data Assimilation Research Section
Data Assimilation Research Testbed Tutorial
Jeffrey Anderson, NCAR Data Assimilation Research Section
Climate and Global Dynamics Laboratory, NCAR
Keith Lindsay and Gokhan Danabasoglu Oceanography Section (CGD/NCAR)
Impact of Traditional and Non-traditional Observation Sources using the Grid-point Statistical Interpolation Data Assimilation System for Regional Applications.
Jeff Anderson, NCAR Data Assimilation Research Section
Parallel Implementations of Ensemble Kalman Filters for Huge Geophysical Models Jeffrey Anderson, Helen Kershaw, Jonathan Hendricks, Nancy Collins, Ye.
Jeff Anderson, NCAR Data Assimilation Research Section
A coupled ensemble data assimilation system for seasonal prediction
Workshop 1: GFDL (Princeton), June 1-2, 2006
Y. Xue1, C. Wen1, X. Yang2 , D. Behringer1, A. Kumar1,
Assimilation of Global Positioning System Radio Occultation Observations Using an Ensemble Filter in Atmospheric Prediction Models Hui Liu, Jefferey Anderson,
Development of an advanced ensemble-based ocean data assimilation approach for ocean and coupled reanalyses Eric de Boisséson, Hao Zuo, Magdalena Balmaseda.
Data Assimilation Initiative, NCAR
WP3.10 : Cross-assessment of CCI-ECVs over the Mediterranean domain
Frank Bryan & Gokhan Danabasoglu NCAR
Presentation transcript:

pg 1 A Loosely Coupled Ocean-Atmosphere Ensemble Assimilation System. Tim Hoar, Nancy Collins, Kevin Raeder, Jeffrey Anderson, NCAR Institute for Math Applied to Geophysics Data Assimilation Research Section Steve Yeager, Mariana Vertenstein, Gokhan Danabasoglu, Alicia Karspeck, and Joe Tribbia NCAR/NESL/CGD/Oceanography AMS – 26 Jan 2011

pg 2 Ocean Data Assimilation Motivation Climate change over time scales of 1 to several decades has been identified as very important for mitigation and infrastructure planning. High fidelity ocean states will be needed by the IPCC decadal prediction program. The ocean plays a crucial role by providing a source or sink (and system memory) for the atmosphere of many quantities, such as heat, moisture, CO 2, etc. Increasing numbers of observations from larger regions of the oceans are making state-of-the-art data assimilation a promising possibility. AMS – 26 Jan 2011 The resulting high-fidelity ocean states are needed. The ensembles provide uncertainty quantification.

pg 3 AMS – 26 Jan 2011 Ensemble state at time of next observation (prior) Ensemble state estimate, x(t k ), after using previous observation (analysis) 1. Use model to advance ensemble (3 members here) to time at which next observation becomes available. Ensemble Filter For Large Geophysical Models

pg 4 2. Get prior ensemble sample of observation, y = h(x), by applying forward operator h to each ensemble member. Theory: observations from instruments with uncorrelated errors can be done sequentially. AMS – 26 Jan 2011 Ensemble Filter For Large Geophysical Models

pg 5 3. Get observed value and observational error distribution from observing system. AMS – 26 Jan 2011 Ensemble Filter For Large Geophysical Models

pg 6 4. Compute the increments for the prior observation ensemble (this is a scalar problem for uncorrelated observation errors). Note: Difference between various ensemble filters is primarily in observation increment calculation. AMS – 26 Jan 2011 Ensemble Filter For Large Geophysical Models

pg 7 5. Use ensemble samples of y and each state variable to linearly regress observation increments onto state variable increments. Theory: impact of observation increments on each state variable can be handled independently! AMS – 26 Jan 2011 Ensemble Filter For Large Geophysical Models

pg 8 6. When all ensemble members for each state variable are updated, there is a new analysis. Integrate to time of next observation … AMS – 26 Jan 2011 Ensemble Filter For Large Geophysical Models

pg 9 Experiment Overview “23 POP 1 DATM” denotes the experiment using 23 POP members and a single “data” atmosphere (CESM framework). “48 POP 48 CAM” denotes the loosely coupled experiment using 48 POP members and 48 consistent but unique CAM atmospheres. Both experiments were conducted for 1998 & The 48 POP 48 CAM experiment was subsequently chosen to produce initial conditions for the IPCC decadal prediction program. AMS – 26 Jan 2011 Compare the effect of using a single atmospheric boundary vs. an ensemble of atmospheric boundaries on ocean data assimilation with POP.

pg 10 Coupled Ocean-Atmosphere Schematic AMS – 26 Jan 2011 Hadley + NCEP-OI2 SSTs Obs used by NCAR-NCEP reanalyses DART/CAM assimilation system CESM1 coupler history files: atmospheric forcing DART/POP assimilation system World Ocean Database Observations CAM analyses: CAM initial files; posterior ensemble mean of state variables prior ensemble mean of all other variables CLM restart files; prior ensemble mean of all variables CICE restart files; prior ensemble mean of all variables CAM state variables = PS, T, U, V, Q, CLDLIQ, CLDICE Prior = values before assimilation (but after a short forecast) Posterior = values after the assimilation of observations at that time POP analyses: temperature, salinity, velocities, surface height

pg 11 AMS – 26 Jan 2011 Atmospheric Reanalysis Assimilation uses 80 members of 2 o FV CAM forced by a single ocean (Hadley+ NCEP-OI2) and produces a very competitive reanalysis. O(1 million) atmospheric obs are assimilated every day. Generates additional ocean spread. Each POP ensemble member is forced with a different atmospheric reanalysis member. 500 hPa GPH Feb

pg 12 FLOAT_SALINITY FLOAT_TEMPERATURE DRIFTER_TEMPERATURE33963 MOORING_SALINITY MOORING_TEMPERATURE BOTTLE_SALINITY BOTTLE_TEMPERATURE CTD_SALINITY CTD_TEMPERATURE STD_SALINITY 674 STD_TEMPERATURE 677 XCTD_SALINITY 3328 XCTD_TEMPERATURE 5790 MBT_TEMPERATURE XBT_TEMPERATURE APB_TEMPERATURE AMS – 26 Jan 2011 World Ocean Database T,S observation counts These counts are for 1998 & 1999 and are representative. temperature observation error standard deviation == 0.5 K. salinity observation error standard deviation == 0.5 msu.

pg POP 1 DATM 1.POP 1-degree displaced pole; 2.23 ensemble members starting from a ‘climatological’ state; 3.Single ‘data’ atmosphere from CORE; 4.DART assimilates observations once per day in a +/- 12 hour window centered at midnight; 5.The CESM framework is responsible for all model advances. The 48 POP 48 CAM experiment differed in that: 1.48 ensemble members starting from a ‘climatological’ state; 2.Atmospheric forcing from the DART/CAM ensemble reanalysis; Guide to the following figures: 1.Ensemble mean 1-day lead forecast difference from observations. 2. is # observations available; +,+ is # assimilated. 3.Obs are rejected if too far from ensemble mean (3 std dev here). AMS – 26 Jan 2011 Experimental Configurations

pg 14 Ensemble Spread for 100m XBT (Expendable Bathythermograph) 1. Spread contracts too much for 23 POP 1 DATM; 2. Using single atmospheric forcing is also part of the problem; 3. Model bias adds to the problem; 4. DART Statistical Sampling Error Correction also helps. AtlanticPacific AMS – 26 Jan 2011

pg 15 Ensemble Spread for Pacific 100m XBT AMS – 26 Jan 2011 Small spread! Twice as much! In general, larger spread rejects fewer obs Spread of the “climatological” ensemble

pg 16 Ensemble Spread for Atlantic 100m XBT AMS – 26 Jan 2011 Small spread! Twice as much! In general, larger spread rejects fewer obs

pg 17 PacificAtlantic 1. Ensemble mean 1-day lead forecast difference from observations. 2. is # observations available. +,+ is # assimilated. 3. Observations are rejected if they are too far from ensemble mean (3 standard deviations here). AMS – 26 Jan m Mooring Temperature RMSE

pg 18 Ensemble mean 1-day lead forecast difference from observations. AMS – 26 Jan m Mooring Temperature RMSE – Pacific POP/CAM as good or better RMSE Similar observation rejection characteristics

pg /3 of the obs are still rejected by 48 POP 48 CAM in the Pacific. 2. Model bias in the thermocline? PacificAtlantic AMS – 26 Jan m Mooring Temperature RMSE

pg 20 AMS – 26 Jan m Mooring Temperature RMSE – Pacific POP/CAM as good or better RMSE Many observations still not assimilated

pg POP 1 DATM Coupled Free Run POP forced by observed atmosphere (hindcast) 48 POP 48 CAM AMS – 26 Jan 2011 Physical Space: 1998/1999 SST Anomaly from HadOI-SST

pg 22 AMS – 26 Jan 2011 Learn about ensemble assimilation and DART tools at: Anderson, J., Hoar, T., Raeder, K., Liu, H., Collins, N., Torn, R., Arellano, A., 2009: The Data Assimilation Research Testbed: A community facility. BAMS, 90, 1283—1296, doi: /2009BAMS2618.1

pg 23 AMS – 26 Jan 2011 End of Presentation, following slides held in reserve.

pg 24 AMS – 26 Jan 2011 Observation Visualization Tools

pg 25 AMS – 26 Jan 2011 Observation Visualization Tools

pg 26 Uses the CESM1 software framework; ocean, atmosphere, and other components communicate through the coupler. A few minor script changes and use of the interactive ensemble capability permit each member of an ensemble of POPs to be forced by a different CAM atmosphere. Once the additional files are staged, the basic implementation is a trivial addition to the run script that invokes the DART system. # # See if CSM finishes correctly (pirated from ccsm_postrun.csh) # # DART assimilation operating on restarts # grep 'SUCCESSFUL TERMINATION' $CplLogFile if ( $status == 0 ) then ${CASEROOT}/assimilate.csh endif AMS – 26 Jan 2011 The CAM-DART-POP Implementation