SERTS-95 measurements of wavelength shifts in coronal emission lines across a solar active region

We used slit spectra from the 1995 flight of Goddard Space Flight Center's Solar EUV Rocket Telescope and Spectrograph (SERTS-95) to measure wavelength shifts of coronal emission lines in the core of NOAA active region 7870 relative to its immediate surroundings (its "edge"). This method circumvents the unavailability of reliable laboratory rest wavelengths for the observed lines by using wavelengths from the edge spectrum as references. We derived the SERTS-95 wavelength calibration from measurements of a post-flight laboratory spectrum containing 28 He II and Ne Ir EUV standard wavelengths known to high accuracy. Wavelength measurements for lines of He I, Ne III, and additional lines of Ne II in the laboratory calibration spectrum provide more accurate values than were previously available, enabling these lines also to serve as future calibration standards. Six solar lines were chosen for this study, namely, He II. at 303.78 Angstrom Fe XII. at 193.51 Angstrom Fe XIII at 202.05 Angstrom, Fe XIV at 211.33 Angstrom, Fe XV at 284.15 Angstrom, and Fe XVI at 335.41 Angstrom. Because these lines are free from known blends in the SERTS-95 spectra and are either intrinsically strong or near the SERTS-95 peak sensitivity, they are our most reliable lines for measuring relative wavelength shifts in the spatially resolved active-region core spectra. The iron ions are the hottest ions ever used for this type of analysis. Ail six lines reveal statistically significant spatial variations in their measured relative wavelength shifts in the active-region core, including mixtures of blueshifts and redshifts (each with maximum values corresponding to relative Doppler velocities similar to 15 km s(-1)), indicating a dynamic, turbulent corona. For each of these lines we calculated weighted-average relative Doppler velocities from the wavelength shifts in the spatially resolved core spectra by weighting the shifts in the individual spatial pixels with their respective measurement uncertainties. This yields velocities of 3.3 +/- 1.1 km s(-1) for He II, 5.2 +/- 1.6 km s(-1) for Fe XII, 0.7 +/- 1.5 km s(-1) for Fe XIII, -2.1 +/- 1.4 km s(-1) for Fe XIV, 1.0 +/- 1.1 km s(-1) for Fe XV, and - 1.1 +/- 0.8 km s(-1) for Fe XVI. We also calculated intensity-weighted relative Doppler velocities from the wavelength shifts in the spatially averaged core spectrum, obtaining corresponding values of 5.8 +/- 0.6 km s(-1), 5.7 +/- 0.9 km s(-1), 0.4 +/- 0.7 km s(-1), - 2.1 +/- 1.0 km s(-1), 0.8 +/- 0.8 km s(-1), and - 1.1 +/- 0.5 km s(-1). Combining the above six lines with several additional ones that are strong enough in both the edge and average core spectra to provide reliable centroid measurements, we find statistically significant net relative redshifts for lines of He II, Fe X, Fe XI, and Fe XII; lines of Fe XIII and Fe XV show no significant shift while lines of Si XI, Fe XIV, and Fe XVI show a small net relative blueshift. Where multiple lines are available for a given ion, the directions (red or blue) and magnitudes (except for Fe XI) of the relative shifts are mutually consistent. The net relative blueshift observed in the hottest active-region coronal lines in our sample, combined with the net relative redshift observed in the cooler active-region coronal lines, suggests a net upflow of heated material cospatially and cotemporally with a net downflow of cooler material.