Towards realtime Space Weather forecasts using CMIT Geomagnetic

  • Slides: 17
Download presentation
Towards real-time Space Weather forecasts using CMIT: Geomagnetic variations and regional K indexes George

Towards real-time Space Weather forecasts using CMIT: Geomagnetic variations and regional K indexes George Millward …with help from: Michael Gehmeyr (LASP), Mike Wiltberger, Wenbin Wang, Pete Schmitt, Richmond, Astrid Maute (HAO) …. and: Terry Onsager, Howard Singer, Chris Balch (SWPC) Art

CISM Knowledge Transfer: Transitioning CMIT for Real-Time Forecasting LFM Real-time Solar Wind Data at

CISM Knowledge Transfer: Transitioning CMIT for Real-Time Forecasting LFM Real-time Solar Wind Data at L 1 Sun Earth Solar Wind Model Real-time Inputs • Initial: Real-time solar wind data from the ACE spacecraft. Lead time: about 1 hour • Future: Solar wind model driven by solar observations Lead-time: 3 to 5 days Magnetosphere Ionosphere Coupler Electric Potential Particle Precipitation Current Geospace Conditions Conductivities Winds TIE-GCM Coupled Magnetosphere Ionosphere Thermosphere model (CMIT 2. 0) • CMIT: First principles, physics-based, 3 D, timedependent model. Consists of a coupling between the Lyon-Fedder-Mobarry MHD Magnetospheric model(LFM) and the NCAR Thermosphere Ionosphere Electrodynamics General Circulation Model (TIE-GCM) Space-Weather Forecast (SWPC) Output Products, User Benefits • Initial: 1 hour forecast of geomagnetic activity, global ionospheric electron content, magnetopause location • Present: Medium resolution mode. Running on “Bluesky”: NCAR IBM Cluster 1600 - 8 proc (Open. MP) • Users: Power utilities, Satellite operators, communication, precision GPS • Future: High resolution fast-mode (MPI). Next generation multi-processor computing resources • Future: 3 day forecast using solar model and observations.

CMIT 2. 5 development on NOAA w. JET computer • w. JET: LINUX Cluster

CMIT 2. 5 development on NOAA w. JET computer • w. JET: LINUX Cluster (1400 cores) • Utilizes Intel compilers • Supports parallel processing via MPI (Not Open. MP) • LFM and LFM-MIX running in parallel on w. JET • Tiegcm v 1. 9 running on w. JET CMIT 2. 5 (LFM MPI - TIEGCM 1. 9) on w. JET 98% built

LFM Magnetosphere Ionosphere Coupler Real-time Solar Wind Data at L 1 Electric Potential Particle

LFM Magnetosphere Ionosphere Coupler Real-time Solar Wind Data at L 1 Electric Potential Particle Precipitation Current Geospace Conditions Conductivities Winds TIE-GCM Space-Weather Forecast (SWPC)

LFM Magnetosphere Ionosphere Coupler Real-time Solar Wind Data at L 1 Electric Potential Particle

LFM Magnetosphere Ionosphere Coupler Real-time Solar Wind Data at L 1 Electric Potential Particle Precipitation Current Geospace Conditions Conductivities Winds TIE-GCM “in house” (NOAA SWPC) Space-Weather Forecast (SWPC)

Real-time Solar Wind Data at L 1

Real-time Solar Wind Data at L 1

LFM Magnetosphere Ionosphere Coupler Electric Potential Particle Precipitation Conductivities Winds TIE-GCM

LFM Magnetosphere Ionosphere Coupler Electric Potential Particle Precipitation Conductivities Winds TIE-GCM

Current Geospace Conditions Outputs/Products

Current Geospace Conditions Outputs/Products

April 2006 Magnetic Storm 2 nd 5 9 April 2006 14 20 th

April 2006 Magnetic Storm 2 nd 5 9 April 2006 14 20 th

April 2006: Magnetometer data

April 2006: Magnetometer data

April 2006: CMIT model output

April 2006: CMIT model output

JN JHorizontal JE

JN JHorizontal JE

Summary / Comments: Exciting initial results: CMIT clearly responding well during storm conditions: Some

Summary / Comments: Exciting initial results: CMIT clearly responding well during storm conditions: Some problems. Future: Running CMIT 2. 5 on NOAA w. JET (soon !!) Further runs needed for calibration/tuning Inclusion of magnetospheric currents (LFM/RCM)