[Paper Review] MOSiC: an analysis tool for IRIS spectral data

R. Rezaei|arXiv (Cornell University)|Jan 16, 2017

Astronomical Observations and Instrumentation1 references3 citations

TL;DR

MOSiC is an IDL-based analysis tool for spectral line fitting in IRIS UV data, employing iterative Gaussian fitting with controlled degrees of freedom to model optically thin and thick lines (e.g., O i 135.6 nm, C ii 133.5 nm, Mg ii h/k, Si iv 140.277 nm). It enables precise Doppler shift, line width, and intensity measurements across chromospheric and transition region lines, with robust noise modeling and velocity correction via O i and Cl i lines, significantly improving spectral analysis accuracy for solar physics research.

ABSTRACT

This is a manual for the MOSiC package. MOSiC is a collection of IDL programs for profile analysis and Gaussian fitting of the Mg II h/k lines along with Gaussian fitting of the C II 133.5 nm line pair, the O I 135.6, the Cl I 135.2, the Si IV 139.7 and 140.3 and the O IV 140.0 nm lines observed with the IRIS near UV and far UV spectrograph. It was tested by analyzing over a hundred different IRIS data sets (quiet Sun, sunspot, ...). It works for off limb data, although it is still experimental. MOSiC analyzes different spectral lines separately and returns line intensity, width, and velocity for each line. A few sample profiles and maps are included in this manual.

Motivation & Objective

• To develop a robust, user-friendly tool for fitting complex UV spectral lines in IRIS data, particularly in the chromosphere and transition region.
• To address challenges in fitting optically thick and thin lines with varying line profiles, including self-reversals and blends.
• To enable accurate measurement of Doppler shifts and line widths by correcting for orbital velocity drift using stable reference lines like O i 135.6 nm and Cl i 135.17 nm.
• To support high-precision analysis of flaring and active regions where line profiles become saturated or highly asymmetric.
• To provide a modular, memory-efficient framework that processes data in raster-by-raster mode, minimizing RAM usage.

Proposed method

• Uses MPFIT for non-linear least-squares fitting with user-defined parameter bounds and degrees of freedom control.
• Applies single, double, quad, or penta-Gaussian models depending on line optical depth and complexity (e.g., C ii 133.5 nm uses up to 13 free parameters).
• Performs iterative fitting with increasing complexity, starting from minimal models and progressing to more complex ones.
• Corrects for orbital velocity drift by using the O i 135.6 nm line as a reference for FUV data and a photospheric line for NUV.
• Estimates UV continuum from line wings and uses this to normalize line profiles and improve fitting accuracy.
• Includes diagnostic tools to visualize fits, check reduced chi-square, and inspect individual pixel profiles via mouse interaction or postscript export.

Experimental results

Research questions

• RQ1 How can spectral lines in IRIS data be accurately fitted when they exhibit complex profiles such as self-reversals or blends?
• RQ2 What is the optimal fitting strategy for lines with varying optical depth (thin vs. thick) across the UV spectrum?
• RQ3 How can orbital velocity drift be effectively corrected using stable, unblended reference lines in FUV and NUV data?
• RQ4 To what extent can MOSiC handle saturated or highly asymmetric line profiles in flaring regions without model failure?
• RQ5 How can the fitting process be optimized to minimize memory usage while maintaining high accuracy in large data cubes?

Key findings

• MOSiC successfully fits optically thin lines like O i 135.6 nm and Cl i 135.17 nm with single-Gaussian models, achieving reliable Doppler shift measurements.
• For optically thick lines such as C ii 133.5 nm, the penta-Gaussian model (13 free parameters) accurately captures the double-line structure and the central Ni ii 133.52 nm line.
• The O i 135.6 nm line provides a stable reference for orbital velocity correction, with velocity maps from O i and Cl i showing strong agreement.
• Reduced chi-square maps help identify problematic fits, especially in flaring regions where line saturation or large Doppler shifts cause model mismatch.
• The tool enables reliable amplitude, width, and center-of-gravity velocity measurements across multiple IRIS lines, including Si iv, Mg ii, and coronal lines.
• Diagnostic tools allow interactive inspection of individual pixel profiles, enhancing confidence in fitting results and error detection.