next up previous contents
Next: Modeling the Three-Dimensional Up: Examples of UVCS Previous: Electron Coronal Density

Small-Scale Temporal Variability in the Corona

A long observation in the ``sit and stare" mode with good spatial and temporal resolution can be used for several studies.

The primary goal is a measurement of the spectral power distribution of turbulence based on the Ly- intensity fluctuations. Initially, it will be important to identify the spatial and temporal scales for comparison with the turbulence observed with EISCAT. Detailed interpretation will be very complex, because the Ly intensity depends on the gas density, the neutral fraction (roughly ) and the velocity (through doppler dimming). Comparison with EISCAT observations, which are only sensitive to density fluctuations, may help to determine nature of the turbulence. At 1 AU, the turbulent spectrum extends from about to 0.1 Hz with a spectral index typically between 1.5 and 1.7. Over much of this range, the compressible component of the turbulence is small, and the turbulence can be considered Alfvenic. The sensitivity of UVCS observations to turbulence has not been worked out in detail, but a streamer has the advantage of relatively high brightness, which improves the sensitivity to rapid, low amplitude fluctuations. Also, a fluctuation of given amplitude is more easily detectable in a localized feature, such as a streamer, than in an observation pertaining to a longer line of sight. Therefore, the intensities in the Table refer to a streamer. We will, however, wish to attempt similar observations in coronal holes and the quiet corona.

UVCS can also make frequent measurements of the intensity of a strong line formed in a small, well-defined structure (e.g. a section of a large loop). Wave dynamics in coronal loops may be described by few spatial modes, no matter how complicated: the remaining `complicating' modes are simply `guided' by the `dominant' ones. An estimate of the number of dominant modes can be found as follows. One takes (many) observations of the fluctuations in a coronal loop ( e.g. fluctuations in some line broadening) at regular intervals. The time series so produced is processed numerically to compute the Lyapunov exponents. From these, one calculates the information dimension (Kaplan and Yorke 1979) of the fluctuations. Starting from the same time series, one can estimate the correlation dimension introduced by Grassberger and Procaccia (1983) which is a lower bound to the information dimension. The estimate of the fractal dimension has a key importance for the theory of wave dynamics in the Solar atmosphere and hence for coronal heating by waves. Routines for the computation of the quantities mentioned above are already available. A typical oscillation of a loop takes place over a period of say 100 sec. We want to cover many of these periods, say 100, with a sufficient resolution of 10 samples per period. This gives some 1000 observations each taken every 10 seconds.

Small-Scale Temporal Variability in the Corona



next up previous contents
Next: Modeling the Three-Dimensional Up: Examples of UVCS Previous: Electron Coronal Density



Peter Smith
Fri Jan 17 12:11:15 EST 1997