The National Earth System Prediction Capability (National ESPC) Project

1. The National Earth System Prediction Capability (National ESPC) Project Daniel P. Eleuterio, Office of Naval Research Jessie Carman, NOAA Office of Ocean and Atmosphere Research Fred Toepfer, NOAA National Weather Service Dave McCarren, Naval Meteorology and Oceanography Command US THORPEX Legacy Planning Meeting June 5, 2014

2. National ESPC Overview • An interagency collaboration, initiated between Navy, Air Force and NOAA and expanded to DoE, NASA, and NSF in 2012, for coordination of research to operations of a National earth system analysis and prediction capability. The original project combined global weather models in an operational synoptic multi-model ensemble under NUOPC. • Seeks improved communication and synergy, for global prediction of weather, ocean, and sea ice conditions at weather to short-term climate variability timescales. • Common prediction requirements and forecast model standards that enable agencies to improve leverage and collaboration. • A national research agenda that will improve prediction across scales from days to decades. • Cooperative focusprojectsto assess predictability of global scale high impact environmental conditions to inform S&T, R&D, and transition to operations. • Towards an multi-model ensemble based air-sea-land coupled global prediction capability http://espc.oar.noaa.gov/

3. Challenges to Achieve a Weather-Ready Nation • Hail, Tornadoes, Tropical Storms & Hurricanes • Winter Storms, Ice • Extreme Heat and Cold • Droughts and Floods • Climate Adaptation • Sea Level Rise • Commerce and Navigation • Aviation Transportation • Food Security • Air and Water Quality • Ecosystem Health • Private Sector and National Security Home Field Emphasis National Weather Enterprise Public Safety and Economic Well-being of Nation Public-Private Partnership

4. Defense Department Challenges – A Global Force for Good Home Field Advantage … at the Away Games We provide worldwide forecasts to support DoD Operations – from the tropics to the poles, and from the depths of the ocean to the edges of space, across the coast to support stability operations, humanitarian assistance and disaster relief.

5. From Sources of Predictability at Weather to Seasonal/ Climate Variability Timescales to Operational Prediction through Global Coupled Model Ensembles

6. NRC Study on Subseasonal to Seasonal Forecasting This study will identify: • Potential sources of predictability and assess their relative value for advancing predictive skill; • Process studies for incorporating new sources of predictability into forecast models; • Opportunities for application and advancement of atmosphere-ocean-land-sea ice coupled models at S2S timescales (a few days to several weeks); • Key observations needed for model initialization and verification of S2S forecasts; • Techniques for uncertainty quantification and verification of probabilistic products; and • Infrastructure requirements for computational, data storage and communication, and visualization techniques needed to make high resolution data assimilating global coupled ensembles an operational reality.

7. Need: Seamless Full Earth System (or at least “lightly seamed”) Partnerships: ESPC NOPP HFIP NMME USGCRP USCLIVAR NUOPC WWRP/WCRP National ESPC

8. Design infrastructure for operational implementation for coupled system Define implementation across operational systems, architecture requirements, cycling setup including DA Coupled Model Development Navy ESPC Highlights - Infrastructure • “Operational Implementation Design” has details on data volumes, resources required, and operational job distributions • “In press” as Naval Research Laboratory Memorandum Report 7320--13-9498 http://www7320.nrlssc.navy.mil/pubs.php search under author Metzger

9. NOAA Next Generation Global Prediction System (NGGPS) Earth System Prediction R2O (draft) Project Information and Highlights Scheduling • Scope: • Improved Model Guidance: Coupled NWP to 30 days • Expand critical weather forecasting research to operations (R2O) • Accelerate development and implementation of improved global weather & ocean prediction models • Improve data assimilation techniques • Improve software architecture and system engineering • Promote hurricane and other high impact forecast models that meet societal requirements • Estimated Benefits: • Address growing service demands • Increased accuracy and lead time for high impact weather forecasts • More reliable probabilistic forecasts • Effectively mitigate economic disruption from hurricanes and other predictable “strong signal” weather phenomena Issues/Risks Increased WCOSS Developmental Computing Capacity Needed Probabilistic operational skill at longer lead times needs to be better quantified and communicated.

10. Air Force Post-Processed Ensemble Products • Unlike Navy and NOAA, USAF does not develop their own global model nor share their forecast fields due to their licensing agreement. Instead they provide joint NUOPC Ensemble processing. • Probability of • Surface wind > 25, 35, 50 kts • Lower and upper level moderate and severe turbulence • Icing at 500, 700, 850, 925 mb • Surface viz < 1, 3, 5 nm • Surface temp < 32, > 90 • Many Precip products and others • All provided to NCEP NOMADS servers as operational fields, 6 hrs time steps out to 10 day leads for all fields. • Plans for longer lead times and higher resolution in 2014/15. Probability of Surface Level Reduced Visibility Probability of Severe Fighter Index of Thermal Stress (FITS)

11. Collaborative Programs Across Scales:Multi-model ensembles • Inter-agency Atmospheric Weather and Coupled Climate R2O Ensembles • Hurricane Forecast Improvement Program(HFIP: 1-7 days) • Providing rapid improvement R2O capability for US (NOAA), • Global (Navy) Tropical Cyclone Track and Intensity • Distributed Production Centers leverage multi-agency resources • National Unified Operational Prediction Capability(NAEFS/ NUOPC: 5-16 days) • Improving medium-range forecasts and probabilities of specific events. • Distributed Production Centers leverage multi-agency missions and resources • Planned extension to 30 days (poss. 45 days) • Increasing resolution to 1/2˚ in 2015 • Incorporating Common Modeling Architecture (ESMF) • National Multi-model Ensemble (NMME: 3-9 months) • Multi-model Climate Ensembles: more accurate than any one model • Distributed Production: leverages multi-agency and • international computer infrastructure and investments (US, Canada) • Currently a Phase II research project through FY14 for higher • resolution output suitable for sub-seasonal updates (Weeks 3-12).

12. DRAFT Interagency National ESPC Fielding Plan (2018) National Global Prediction Needs Forecast Lead 0-7 Days Forecast Lead 8-14 Days Forecast Lead 15-30 Days Forecast Lead 31 – 90 Days Forecast Lead 3 – 18 Months Annual to Decadal Strategic Planning, Environmental Stress Instability, Arctic Sea Lanes, Ship and Land Transportation Infrastructure Management, Agriculture/Fisheries Planning & Ecosystem Management, Water Resource Management, Energy Sector Planning . . . Boundary conditions for TC Track & Intensity, TC genesis, Hi Res Littoral air/ocean operations, EM/EO/NBC prediction Drought/Flood, Heat Wave/Freeze Prediction, Storm Surge/Inundation/Erosion, Beach Warnings, Safety/Emergency/Public Health Operations, Ship & Aircraft Routing. . . Navy and DoD Capability TC-COAMPS Others NAVGEM/HYCOM/CICE/WW-3 NAVGEM Ensemble NUOPC/NAEFS Ensemble DOD Applications of NOAA managed Seasonal Ensemble Prediction Static Climatology/Reanalysis based on past environment “Climatology Products” HFIP Ensemble NUOPC Ensemble NOAA Capability Climate Fcst System (CFSv2/3) National Multi-Model Ensemble (NMME) HWRF GFDL GFS CFS-R, HURDAT, etc. GFS, HYCOM, WW-3 GEFS Ensemble Multi-Model Ensemble

13. The call for Extended Range Products in the S2S Gap(Federal/Operational, Credible, Reliable, Calibrated, and Probabilistic) • When initialized anywhere in phase space, the solution collapses toward this attractor. However, behavior of the trajectory is extremely sensitive to the initial conditions. -- Ed Lorenz (1963) • Every day meteorologists predict next week’s weather…..Research is underway to develop models that will help them make predictions on an even longer timescale, seasonal forecasts of monsoon rains for example. --Tim Palmer, A weather eye on unpredictability (1991) • Nearly 25 years later are we as far along as we should be?

14. backup