University of Colorado at Boulder University of Colorado CU Home Search A to Z Index Map
Laboratory for Atmospheric and Space Physics

Cassini UVIS Spectral Analysis

Here is a link to the UVPapersTable.

The paper about the Cassini UVIS instrument is Esposito et al. (2004), Space Science Reviews, 115, 299-361,

The LASP group has used CHIANTI for their spectral analysis of the Cassini UVIS spectra. Documentation about how we simulate a spectrum using CHIANTI is provided here: CHIANTI_Emiss_Documentation

All the necessary IDL code to simulate a spectrum, with the exception of the CHIANTI database version 8.0.7 and accompanying routines, can be found here:

The CHIANTI database version 8.0.7 and accompanying routines can be found here: CHIANTI

Andrew Steffl’s PhD Thesis (U. Colorado 2005) involved a significant amount of analysis of Cassini UVIS data from the Io plasma torus:

Steffl, Stewart, Bagenal, (2004a). Cassini UVIS observations of the Io plasma torus. I: initial results, Icarus, 172, 78-90

Steffl, Bagenal, Stewart, (2004b). Cassini UVIS observations of the Io plasma torus. II: radial variations, Icarus, 172, 91-103

Steffl, Bagenal, Stewart, (2006). Cassini UVIS observations of the Io plasma torus. III: Observations of temporal and azimuthal variability, Icarus, 180, 124-140

Steffl, Bagenal, Stewart, (2008). Cassini UVIS observations of the Io plasma torus. IV: Modeling temporal and azimuthal variability, Icarus, 194, 153-165

Recent Analysis can be found in our paper:

Nerney, E. G., Bagenal, F., Steffl A. J. (2017). Io Plasma Torus Ion Composition: Voyager, Galileo, Cassini, J. Geophys. Res., Jan 2017, DOI: 10.1002/2016JA023306

To investigate the implications of changes in the atomic data from CHIANTI 4.2 (as used by Steffl et al. 2004b) to CHIANTI 8.0, we fit the Cassini UVIS spectra for torus ansa distances of 6.4 RJ and at 7.85 RJ as shown in Figure 4. When forward-fitting the spectra we need to assume a full-width half-maximum (FWHM) spread of each emission line. A FWHM of 3 A is quoted by Steffl et al. (2004a) while a value of 4.7 A was used by Yoshioka et al. (2011). We found 3 A to be too narrow and that a FWHM of 4.47 A produced the best fit to the spectrum.

Overall we are able to find a reasonably good fit to the spectrum but a few S++ features are under fit. Specifically, model emissions for SIII 680, 702, and 1729 are noticeable below values in the observed spectrum. The quality of the fit is similar to that shown in Figure 4 of Steffl et al. (2004b). It is not clear if the quality of the fit is limited by calibration, averaging of the emissions over a line of sight, the atomic data, or, most likely, some combination thereof. New atomic data for SIV(S+++) (Del Zanna et al. 2016) will be added to the next version of CHIANTI that may help the overall match to the data.

Screen Shot 2017-02-07 at 1.00.07 PM

Screen Shot 2017-02-07 at 1.00.59 PM

The determination of the composition of the oxygen species is complicated by the fact that the 833—834 A feature is a combination of O+ and O++ emission. The FUV component of the Cassini UVIS spectrum includes a couple lines of O++ emission at 1661 and 1665 A. While these lines are weak, they allow at least an upper limit to be set on the abundance of O++

Screen Shot 2017-02-07 at 1.01.52 PM

To improve the signal-to-noise of the O++ signature we do a 5-bin running average of the UVIS, effectively over a radial distance of about 1 RJ. We start by fitting the whole spectrum to determine electron temperature and sulfur ion column densities while holding the electron density fixed (according to the profile described above) and not fitting the oxygen line. We then hold the electron temperature and sulfur composition constant at these best-fit values and fit the O++ feature at 1661/1666 A to find an upper limit on the amount of O++. We then hold everything except O+ fixed at the best fits just found while we fit the 833/834 blended feature. This allows us to constrain the O++ so the least squares fitting procedure does not get lost in parameter space.

The table above shows the mixing ratios of the sulfur and oxygen species for our analysis using CHIANTI 8.0 and the process described above compared with what Steffl et al 2004 found. The ion composition is basically the same derived by both versions of CHIANTI for both spectra. Both analyses show ~12% increase in charge state but little change in the total O/S content with radial distance.

In a recent paper which focuses on the torus emissions observed by the Cassini UVIS at the distance of Europa’s orbit (9.4 RJ), Shemansky et al. (2014) find a very different composition (very high O++) to both Steffl’s papers as well as to the original Shemansky (1988) analysis of the Voyager data. Until the flat-field, calibration and atomic data used in the Shemansky et al. (2014) study are published, it is difficult to compare with the current analysis.