1. NOAA Space Weather Prediction Center: Services, Needs, Directions for the Future Christopher C. Balch NOAA Space Weather Prediction Center CCMC R2O Workshop Goddard Space Flight Center October 15-18, 2007

2. CCMC R20 Workshop2 Outline  Hopes, goals, objectives for this workshop  Users of SWPC services  Current products & services  Current tools & data  Performance metrics: current forecasts  Service gaps  Plans – directions for the future  Conclusion Acknowledgments: Singer, Onsager, Murtagh

3. CCMC R20 Workshop3 Some Objectives  Growing national need for space weather services  NOAA resources – level funding  Modernization: depends external partnerships:  Directed Research: NASA, NSF, DOD  Model Validation/Verification: CCMC, USAF  NOAA/SWPC refocusing resources  Understanding user needs  Priority for model transition  Goal: define clearly problems we need to solve

4. CCMC R20 Workshop4 Impact AreaCustomer (examples)Action (examples)Cost (examples) Radiation dose (dose limits & health risks, possible acute effects for space exploration missions) NASA exploration ISS Space tourism Airline crews/passengers Postpone activities Seek shelter Re-route flight paths Safety Issue Health risks Spacecraft (Individual systems to complete spacecraft failure; comm, drag, and radiation effects) Lockheed Martin Orbital Boeing Space Systems Loral NASA, DoD Postpone launch In orbit - Reboot systems Turn off/safe instruments and/or spacecraft Maintain orbit Loss of spacecraft ~$500M Commercial loss exceeds $1B Worst case storm - $100B Electric Power (Equipment damage to electrical grid failure and blackout conditions) U.S. Nuclear Regulatory Commission N. America Electric Reliability Corp. Allegheny Power New York Power Authority Adjust/reduce system load Disconnect components Postpone maintenance Estimated loss ~$400M from unexpected geomagnetic storms $3-6B loss in GDP (blackout) Airlines (Communications) (Loss of flight HF radio communications) United Airlines, Continental, Lufthansa, Korean Airlines NavCanada (Air Traffic Control) Divert polar flights Change flight plans Change altitude Select alternate comms Cost ~ $100k per diverted flight $10-50k for re-routes Surveying & Navigation (Use of magnetic field or GPS could be impacted) FAA-WAAS Dept. of Transportation BP Alaska and Schlumberger Fugro Chance – offshore oil rig Postpone activities Redo survey Use backup systems From $50k to $1 mil daily for single company User Community

5. CCMC R20 Workshop5 Recent Trends  Steady overall growth of users  Fastest growing user areas: Polar Aviation & GPS

6. CCMC R20 Workshop6  Drivers for Polar Aviation AFlight time reductions of 1 to 3 hours AAbsence of turbulence and convection AAvailability of modern aircraft with 6k to 9k mile range AEconomic growth of China and India  Drivers for GPS market ADeep-sea drilling ASurveying AFAA navigation systems ADOD operations AMining & Farming operations Recent Trends

7. CCMC R20 Workshop7 Current Products & Services Solar flare/proton event probability forecast III. Event Probabilities 12 Oct-14 Oct Class M 01/01/01 Class X 01/01/01 Proton 01/01/01 PCAF Green

8. CCMC R20 Workshop8 Current Products & Services 10.7 cm radio flux forecast IV. Penticton 10.7 cm Flux Observed 11 Oct 069 Predicted 12 Oct-14 Oct 069/069/069 90 Day Mean 11 Oct 068

9. CCMC R20 Workshop9 Current Products & Services Geomagnetic probability and A-index forecast V. Geomagnetic A Indices Observed Afr/Ap 10 Oct 001/003 Estimated Afr/Ap 11 Oct 002/005 Predicted Afr/Ap 12 Oct-14 Oct 005/005-005/005-005/005 VI. Geomagnetic Activity Probabilities 12 Oct-14 Oct A. Middle Latitudes Active 10/05/05 Minor storm 01/01/01 Major-severe storm 01/01/01 B. High Latitudes Active 15/10/10 Minor storm 01/01/01 Major-severe storm 01/01/01

10. CCMC R20 Workshop10 Alerts, Warnings, Watches  Alert - Observed condition  X-ray, proton, radio, geomagnetic  Warning – Imminent condition – high probability  Proton, Geomagnetic  Watch – Elevated probability for condition  Geomagnetic

11. CCMC R20 Workshop11 External models  D-Region Absorption  STORM time empirical ionospheric correction model  U.S. Total electron content map  Costello Kp  Wang-Sheeley  Relativistic Electron Forecast Model

12. CCMC R20 Workshop12 Tools & guidance  Flare prediction  Active region monitor  Statistics based on spot class  Historical tracking of Region parameters  Proton event prediction  Proton model (statistical)  Database of proton events  Geomagnetic forecasts  Recurrence  Geomagnetic climatology  Empirical prediction rules

13. CCMC R20 Workshop13 NOAA POES NOAA GOES NASA ACE ESA/NASA SOHO L1 ACE (NASA) –Solar wind speed, density, temperature and energetic particles –Vector Magnetic field SOHO (ESA/NASA) –Solar EUV Images –Solar Corona (CMEs) GOES (NOAA) –Energetic Particles –Magnetic Field –Solar X-ray Flux –Solar EUV Flux –Solar X-Ray Images POES (NOAA) –High Energy Particles –Total Energy Deposition –Solar UV Flux Ground Sites –Magnetometers (NOAA/USGS) –Thule Riometer and Neutron monitor (USAF) –SOON Sites (USAF) –RSTN (USAF) –Telescopes and Magnetographs –Ionosondes (AF, ISES, …) –GPS (CORS) Key Data Sources STEREO (NASA) –Solar EUV Images –Solar Corona & Heliosphere (CMEs) –In-situ plasma & fields –In-situ energetic particles –SWAVES

14. CCMC R20 Workshop14 Forecast Verification & Challenges  Verification Web Page  Flare/Proton probabilities  Lack of vector magnetograms  Empirical concepts need to be tested  No physical models in operations  Proton event warnings  Empirical concepts need to be tested  Very difficult physics problem  Geomagnetic forecasts  Arrival time problem  Uncertainty in speed and direction  Influence of the ambient solar wind  Uncertainty in v B s

15. CCMC R20 Workshop15 Forecast Verification Example  SEP probability prediction model  Requires probabilistic verification  Metrics/Attributes: (Hsu & Murphy, 1986) MS error = 0.0249 RMS error: 0.16 MSE = MSE * + REL – RES MSE * = 0.0324 REL = 0.0007 RES = 0.0082  Target to beat:  Smaller MS error  Better reliability  Better resolution

16. CCMC R20 Workshop16 Forecast Verification Example  Categorical Forecasting: Yes/No  Contingency Table Verification  Key Measures  Probability of Detection (POD)  False Alarm Rate (FAR)  Heidke Skill Score (HSS) Event Observed Event Forecast YesNo YesABnw NoCDnn npncN  Results depend on a threshold  FAR & POD decrease as threshold increases  HSS maximum 0.48  Target to beat at maximum HSS:  POD 57%  FAR 55%  Study in progress: HSS 0.54  POD 51%, FAR 39%

17. CCMC R20 Workshop17 Service Gaps – SWPC priorities  Radiation safety  NASA/SRAG, space tourism, high fliers: all clear forecasts, longer lead times  Airlines: working on standards, operating procedures, regional information  SEP spectra, duration, fluence predictions  GIC – regional maps with nowcast & forecast  Polar Route Airline Comm – regional maps with nowcast  GPS  Regional products with more direct application  Forecasting the ionosphere and its impact on GPS  Possible data gaps in the future:  Solar wind data (ACE lifetime)  Coronagraph data (SOHO/LASCO lifetime)

18. CCMC R20 Workshop18 Challenges, Needs, Tools, Plans Challenges  Integrating large volume of data into a consistent physical picture  Visualization of Active Region Magnetic fields – consistent with the data  Visualization of disturbances in interplanetary space  View from above the ecliptic tracking one or more ICMEs  View of High Speed Streams and the Sector Structure  Visualization of the state of the magnetosphere Plans  Focus: model transition  Prioritization: based on user requirements  Coordination: Must leverage directed research funded through NASA/NSF/DOD  Validation: ‘How good is it ?’ – a must do

19. CCMC R20 Workshop19 Validation Example Costello Prediction Model  Point prediction not sufficient  Error bar – 50% verification interval  The key user question: Given the prediction, what is the distribution of the observations ?  Don’t need perfection !!!  Do need to characterize the uncertainty

20. CCMC R20 Workshop20 Validation Example U.S. Total Electron Content Model  First map: TEC  Second map: standard deviations, i.e. model uncertainty

21. CCMC R20 Workshop21 Summary  Growing national need for Space Wx Services  Way forward for NOAA/SWPC:  We want to focus our efforts on model transition  Requires leveraging work of external partners  Requires communicating problems that need to be solved  Specific Areas:  Improving current forecasts: flares, proton events, geomagnetic storms  Service Gaps: Radiation, Airline communications, GIC, GPS users  Handling the data: visualizations, assimilation  Need to measure improvements objectively  Look forward to working together with CCMC on model validation