GONG Measurements Preeruptive signatures Frank Hill Rudi Komm

  • Slides: 13
Download presentation
GONG Measurements – Pre-eruptive signatures Frank Hill, Rudi Komm and the GONG Team fhill@noao.

GONG Measurements – Pre-eruptive signatures Frank Hill, Rudi Komm and the GONG Team fhill@noao. edu

What is GONG? • The Global Oscillation Network Group • Deployed in 1995 to

What is GONG? • The Global Oscillation Network Group • Deployed in 1995 to continually observe solar oscillations and infer internal solar structure and dynamics (helioseismology) • Six instruments located around the world

What does GONG observe? • Full-disk Doppler velocity, line-of-sight magnetic field, and intensity •

What does GONG observe? • Full-disk Doppler velocity, line-of-sight magnetic field, and intensity • Uses Ni I 676. 8 -nm spectral line • Solar image is 800 x 800 pixels (2. 5” pixels) • One data set every 60 sec at each site • Semi-automated operation

GONG temporal coverage 1995 2007 Overall average duty cycle: 0. 849 Last year: 0.

GONG temporal coverage 1995 2007 Overall average duty cycle: 0. 849 Last year: 0. 893 No day without data since July 2001 Data loss due to instrumental problems is ~0. 02

GONG Space Weather Products • High-cadence (60 sec) surface magnetic field measurements and field

GONG Space Weather Products • High-cadence (60 sec) surface magnetic field measurements and field extrapolations provide rapid changes in the field geometry • Images of far-side magnetic field from helioseismology show large active regions – see poster S-14 by Irene Gonzalez-Hernandez • Helioseismology ring diagrams can measure subsurface vorticity, which is an excellent flare indicator, and may be able to predict the time of flare onset • Considering adding 1” pixel high-cadence (15 -60 sec) full-disk H imaging

Helioseismology • Sun is filled with 5, 000 distinct sound waves • Sound is

Helioseismology • Sun is filled with 5, 000 distinct sound waves • Sound is trapped in a cavity defined by the internal thermal structure • Each wave samples a different range of depths • Can “invert” wave properties to infer internal conditions, such as bulk plasma velocity

Ring Diagrams - I • Three-dimensional power spectrum of solar oscillations • Subsurface flow

Ring Diagrams - I • Three-dimensional power spectrum of solar oscillations • Subsurface flow displaces rings No tracking for solar rotation With tracking

Ring Diagrams – II • Perform analysis over solar disk • Invert for horizontal

Ring Diagrams – II • Perform analysis over solar disk • Invert for horizontal flows as function of location and depth On a single day, can reach ± 60° in heliocentric longitude and latitude.

AR 10069 Flows below strong flare producers AR 10486

AR 10069 Flows below strong flare producers AR 10486

Flare activity, vorticity and B Each point identifies the observed vorticity and surface magnetic

Flare activity, vorticity and B Each point identifies the observed vorticity and surface magnetic field strength B for a unique active region. Filled symbols: multiple flares Open symbols: single flare Sample approximately 700 active regions covering 20012004

Flare occurrence probabilities as a function of vorticity and surface magnetic field The vorticity

Flare occurrence probabilities as a function of vorticity and surface magnetic field The vorticity and surface B measurements are placed in bins, each containing 28 unique active regions. The numbers in the bins are the percentage of active regions with at least one flare with a magnitude above the indicated class. Colors indicate the probability levels in steps of 10%, white is 0%.

Temporal behavior of kinetic helicity before a flare Time history of kinetic helicity (KH)

Temporal behavior of kinetic helicity before a flare Time history of kinetic helicity (KH) below 9 solar areas preceding the Halloween flare. The flaring region was located in the ellipse. The KH was anomalously high before the flare, then abruptly decreased at the time of the flare. The KH may be a predictor of strong flares. Different colors indicate different depths.

Next steps • Fully test added value of combining vorticity with surface magnetic field

Next steps • Fully test added value of combining vorticity with surface magnetic field measurements • Beginning collaboration with KD Leka, G Barnes and A Reinard to apply discriminant analysis and investigate correlation with spot complexity • Start statistical analysis of temporal predictor • Develop automated pipeline at remote sites to estimate the vorticity below active regions on an 8 -hour cadence – would produce six observations per 24 hours • Work with NOAA towards an operational product • GONG facing funding challenges, uncertain future – could result in loss of a useful space weather asset