Ultraviolet Coronagraph Spectrometer (UVCS/SOHO)

Observations of the April 21, 2002 event




UVCS images represent observations taken at 1.63 solar radii from 00:45 UT to 01:55 UT. Each UVCS image is a time sequence of 32 exposures each lasting 120 seconds. The horizontal dimension is the length of the slit (40.6 arcmin) which covers position angles from 232 to 302 degrees. At the start of observations, the bright, relatively stable corona above this active region was present until the CME started to pass at 01:16 UT. The corona was disrupted by the CME (see near the center of the images). The data indicate that this CME was composed of hot material, compared to CMEs resulting from prominence eruptions. Spectral lines from ionized gas hotter than the average corona (e.g., [Fe XVIII] 974 A, 6 million degrees) were present. The [Fe XVIII] emission may originate in a current sheet or in a hot, compressed region where plasma flowing along the current sheet encounters a magnetic flux rope.




The figure is a composite image taken by three SOHO instruments at the time of the 21 April 2002 solar flare. The green image of the solar disk at the time of the flare was taken by EIT in light emitted by iron ions at about 2 million degrees Kelvin. The blue/white image of a coronal mass ejection associated with the flare was taken by LASCO in visible light. UVCS images of the region outlined in white were made in light from four ultraviolet emissions, each showing threads of ejected material at very different temperatures above the flare. The `images' are constructed from a time sequence of observations as the ejected material moved through a region located 0.63 solar radii above the flare site. From top to bottom the UVCS images are total intensities from the emission of iron ([Fe XVIII] 974 A), silicon (Si XII 499 A), oxygen (O VI 1032,1037 A), and neutral hydrogen (H I Lyman alpha 1216 A). These emissions tend to form in regions where the gas is at 6 million degrees K, 2 million degrees K, 300 thousand degrees K, and 20 thousand degrees K, respectively. Spectroscopy can be used to probe the extreme conditions during a solar flare where non-thermal processes are thought to be involved in producing the observed emissions. This composite image was produced by Alexander Panasyuk (SAO).




The UVCS entrance aperture was located at 1.63 Solar Radii, directly above Active Region 9906 during the X1 flare and fast CME. The image shows changes to the OVI (1032 A) emission line at three times during the observations. The vertical axis corresponds to the spatial coordinate along the entrance slit (in arscsec). The spectrograph slit crosses the expanding CME bubble, recording line-of-sight velocities up to 750 km/s toward the Earth and 500 km/s away from the Earth. The predominance of blue-shifts over redshifts suggests that this event was skewed in an Earthward direction in spite of its location almost at the limb.





Click on the image to view the movie

The Doppler shifts provide the plasma velocities along the line of sight. To construct a 3D representation of the CME we assume that the velocity in the plane of the sky for each pixel along the slit in each exposure remains constant at subsequent times. To show the evolution in 3D, the movie shows the structure as viewed from 3 vantage points; the Earth, 1 AU north of the Sun, and 1 AU West of the Sun. The view from the Earth should resemble images from LASCO, except that by choosing the O VI emission lines we select plasma at temperatures or 100 thousand to milion degrees K. The view from the West shows the higher speed material moving toward Earth and the denser material moving away.




This plot shows the intensities of several emission lines averaged over a 168" band centered on PA=262 degrees, the position where the transient [Fe XVIII] emission appeared. The O VI and Si XII emission begin to decline about 10 minutes before the [Fe XVIII] appears. O VI continues to decline smoothly through the [Fe XVIII] peak, while Si XII shows a modest bump at the time of the peak. In the context of flux rope models, this suggests that the [Fe XVIII] region was located near the bottom of the flux rope or perhaps inside it.






This plot shows the intensity distributions of several lines along the UVCS slit before the eruption. Peaks located about 8' north and 2' south of PA=262 degrees are present in the O VI and Si XII lines. However, a single peak located between them is seen in Ly beta. This indicates a streamer whose core is depleted in oxygen and silicon, as has been seen for quiescent streamers. The structure has been as a closed field core with open field legs, in which the metals in the core are depleted by gravitational settling. The CME first appeared at the location of the southern leg and quickly spread in both directions along the slit. Very faint [Fe XVIII] emission was present before the eruption, primarily in the streamer core. It is unusual for [Fe XVIII] to be detectable at this height, but not unprecedented. This indicates that the streamer core was unusually hot. The pre-CME streamer has a density of 2-3x10^7 cm^-3 based on the O VI doublet ratio. The emission measure is fairly flat at 3x10^24 cm^-5 in each interval of 0.1 in log T from 6.0 to 6.4. The [Fe XVIII] feature has a similar density if it is roughly cylindrical. It's average temperature is at least log T=6.5 based on the [Fe XVIII]/Si XII ratio.






This plot shows the intensity of the O VI 1032 A line as a function of time, along with the O VI 1037 A line multiplied by factors of 2 and 3. These intensities are averages over a 13 arcminute region spanning the two peaks seen in the O VI brightness distribution along the slit. Before the eruption the lines are in a 3:1 ratio, a value typically observed in streamers and indicative of approximately equal contributions of collisional excitation and scattering of photons from the solar disk to the O VI line intensities. As the eruption progresses, the ratio quickly drops to 2:1, indicating pure collisional excitation. The radiative scattering component disappears due to Doppler dimming at high outflow speeds.






This plot shows the velocities of the red-shifted and blue-shifted components of the O VI 1032 A line as a function of time at two positions along the UVCS slit. The line profiles were averaged over a 3.5 arcminute region. The solid line shows velocities at the position where the CME first appears and where the [Fe XVIII] feature appears. The dashed line shows velocities at a position about 3 arcminutes to the north. In general, the red-shifted features reach a maximum speed 5-10 minutes after they appear, then shift back toward rest velocity. Blue-shifted features, on the other hand, seem to reach a high speed, then fade away.




Original data files:

The UVCS/SOHO Data Analysis Software (DAS), calibration and help files are available from the UVCS webpage under Get Involved/ Access to Data and Software.