Chapter 8
Retrieving and Analyzing FUSE Data

This chapter highlights the locations where FUSE data can be obtained and also provides brief descriptions of available tools that can be useful for analyzing and interpreting FUSE spectra. More information can be found on the CalFUSE Page. Lists of computer software, spectral atlases, and papers dealing with particularly difficult FUSE analysis problems are also given.

The contents of this chapter are relevant to all future FUSE users regardless of their expertise level ("Casual", "Intermediate" or "Advanced").

8.1 Retrieving FUSE Data

FUSE data are available through three different on-line interfaces. These are:

• The Multimission Archive at STScI (MAST) http://archive.stsci.edu/.
• The Canadian Astronomy Data Centre www2.cadc-ccda.hia-iha.nrc-cnrc.gc.ca/cadc/
• The Institut d'Astrophysique de Paris (IAP) http://fuse.iap.fr/

Each of these sites has its own strengths and is worth examining. In each case, their pages, forms and search procedures are well documented, although MAST maintains far more supporting material than the others. A detailed description of the files and file extensions archived at MAST is given in Chapter 4 of this document. The users are encouraged to refer to Table 4.1 prior to downloading data from the MAST interface.

8.2 Analyzing FUSE data

Because FUSE data are standard FITS files, they can be read and displayed with all standard astronomical software packages. There are also several customized FUSE analysis packages written in IDL and in C that will assist the user in displaying and analyzing FUSE data. These tools, and their accompanying documentation, are available from the MAST or IAP web sites and are only briefly described here.

8.2.1 IDL Analysis Packages

Considerable effort has been put into developing IDL software for analyzing FUSE data. In this section, we describe a few of the well-developed and documented packages and procedures.

First, Don Lindler has written a comprehensive suite of IDL routines customized to handle FUSE data. These are called ltools. The complete package and some documentation are available from the MAST FUSE web pages. The following is an outline of the capabilities of ltools, and some of its major elements.

cf_edit.pro

allows the user to examine, manipulate, and combine the intermediate data files (IDFs) generated by CalFUSE. Users without access to IDL can use the "Virtual Machine" version of the program. The best introduction to the IDF files is probably the cf_edit.pdf, the User's Guide.

fuse_scan.pro

allows interactive manipulation of FUSE images and time-tag files. The data are displayed as an image and can be overlayed with an approximate wavelength scale. Statistics can be computed for the whole image or a selected region, and the program has a variety of tools to manipulate time-tag data. (You can even make a movie!) The program has tools to fit Gaussian emission or absorption features or to produce surface or contour plots of small regions of the image. Documentation is available in fuse_scan_reference.pdf.

fuse_analysis.pro

is an interactive viewer for calibrated FUSE data. The program can simultaneously display spectra from all four detector segments for a single exposure. In addition to plotting the calibrated results, it will display regions of the raw image from each detector.

fuse_register.pro

is an interactive tool for registering, scaling, and coadding calibrated FUSE spectra. The routine allows you to register and coadd multiple exposures for the same channel, or to register, scale, and merge data from multiple channels.

lineplot.pro

is a general-purpose plotting tool used by the FUSE IDL data-analysis widgets. It allows the user to overplot up to 10 plots. It can be used as a stand-alone widget or called from another widget.

line_edit.pro

is a general purpose tool used by the FUSE IDL data analysis widgets for creating data masks of 1-dimensional data. It can be used as a standalone widget or called from another widget.

xgaussfit.pro

is an interactive least-squares multiple Gaussian fitting routine.

line_norm.pro

is an interactive tool for continuum normalization of a spectrum. The tool can be used as a stand-alone routine or called from another widget.

line_eqwidth.pro

is a spectral-feature analysis tool using the algorithms of the International Ultraviolet Explorer Feature routine. It can be used as a stand-alone routine or called from another widget.

NOTE: Don Lindler's routines have been updated to work with IDL 6.3. If you are unable to display spectra when running fuse_register or cf_edit on a Mac or PC, try installing the latest versions of the astron and ltools libraries. Another trick is to add the following lines to your idl_startup.pro file.

;  Let's deal with the 24-bit display issues.
device, true_color=24
window, /free, /pixmap, colors=-10
wdelete, !d.window
device, decomposed=0, retain=2, set_character_size=[8,10]
device, get_visual_depth=depth

;  Print helpful information at login.
print, ' '
print, 'Display depth: ' + strtrim( depth, 2 )
print, 'Color table size: ' + strtrim( !d.table_size, 2 )
print, ' '
$pwd
$hostname
print, ' '

Some additional utility routines are:

readit.pro

is a utility program which will read FUSE format raw FITS files for the purpose of turning them into 2D images for display. It automatically determines the file type from the file header. It is used by some of the other programs provided here.

plotrate.pro

reads in a raw ttag FITS file and plots the count rate as a function of time.

showdetector.pro

reads in a FUSE raw FITS file and makes a gray-scale image plot of the detector segment for printing.

allsegments3.pro

plots extracted spectra from all 4 detector segments onto a single sheet of paper.

wfusehist.pro,

and wfusettag.pro write FUSE HIST or TTAG files.

Here are some additional contributions.

idf_shift.pro

was written by John Grimes and used to modify the wavelength arrays of the LiF and SiC target spectra within an individual IDF. It is useful, for example, if you want to align the spectra from individual exposures to a common wavelength scale before combining their IDFs. The program, idf_shift.pro, can be called from the command line using the Perl script idf_shift.pl.

idf_jitter.pro

is an IDL script that identifies times when a bright target was out of the aperture. It uses the count-rate array in the timeline table of the IDF to construct a count-rate histogram, then rejects times when the count rate was more than 3σ below the mean. For such times, the jitter bit is set in both the timeline table and the photon list.

[H2ools] is a package of IDL routines written by Steve McCandliss for modelling molecular hydrogen absorption. His page provides all sorts of tools and support materials, and the programs are documented in McCandliss (2003).

[XIPLOT] The spectral manipulation tool XIPLOT is an IDL program available at the French FUSE site.

A final IDL tip to those users wishing to read FUSE data on their own: when using the IDL Astronomy User's Library routine mrdfits.pro, the keyword fscale should be set to automatically scale numbers to floating point format when needed.

8.2.2 C Analysis Packages

FUSE Tools in C is a package of data analysis tools distributed along with the CalFUSE pipeline. They come with the document FUSE Tools in C, which reviews each routine. A subset of these programs is designed specifically for the manipulation of IDF files, and is described in the IDF Cookbook, which is available in both PDF and HTML versions on the FUSE Tools web page http://archive.stsci.edu/fuse/analysis/toolbox.html).

These tools have the flexibility to perform many common tasks, including trimming, modifying and manipulating large sets of multiple data files. For example, when combining spectra of bright targets, the goal is usually to to maximize spectral resolution, so it is important to align the spectra from individual exposures precisely before combining them. The FUSE Tools in C package includes routines to accomplish this task. In contrast, when combining spectra of faint targets, the goal is typically to maximize the fidelity of the background correction by increasing the signal-to-noise ratio on unilluminated regions of the detector. To accomplish this, the IDF files must be combined before extracting the spectra. This is easily done with the Tools, and there are detailed instructions in the IDF Cookbook available on the CalFUSE Page. The tools also have routines that enable one to divide an IDF file into multiple time segments in order to examine time dependence.

In a few cases, it may be necessary to reprocess the data using the CalFUSE pipeline. The CalFUSE source code and documentation for installing and using it are available at http:/archive.stsci.edu/fuse/analysis/calfuse.html.

8.3 FUSE Data Quality Check-List

Before performing detailed science investigations with FUSE data, it is recommended that the user (regardless of expertise level) go through the sanity checks given below to avoid erroneous results and interpretations.

• Compare SiC data with overlapping LiF data or night-only SiC data to identify scattered solar light.
• Compare SiC and LiF data with corresponding night-only data, when present, to understand and identify airglow contamination.
• Examine IDF files to search for dead zones, remaining scattered light, background contamination (due to the presence of an extended source).
• Compare spectra from multiple segments to identify exposures where flux was lost due to "worms" or target motions.
• Check count-rate plot (*rat.gif) to understand flux loss to due target motions.
• Examine HIST data for evidence of fixed-pattern noise.
• Check background subtraction accuracy: verify that the flux at the center of a damped line reaches zero.
• To ensure highest spectral resolution when combining extracted spectra, cross-correlate individual exposures with unblended, sharp and solitary lines in the wavelength range of interest. Don't combine spectra from different channels if highest spectra resolution is important.
• To mitigate the effects of the walk and the non-linearity of the wavelength solution, limit the range over which spectral line modeling is performed to 1-3 Å maximum. Produce as many such small windows as required to fit the lines of interest. Beware of detector segment 2A for data obtained after mid-2002, especially in the vicinity of the airglow lines noted in Section 9 of the Instrument Handbook (2009). See Figure 7.7 for an example of the effects of localized gain sag induced by airglow exposure in segment 2A.

8.4 Additional Analysis Aids and Atlases

There are a number of additional resources that are useful for analyzing FUSE data. These include line lists, spectral atlases, and a few other resources.

8.4.1 Atomic and Molecular Line Data

Don Morton has published an atlas of atomic data, complete with finding lists. These are available in the publication by Morton (2003). Because molecular hydrogen absorption is so prevalent in FUSE spectra, it is also important to be aware of the following references to the basic H₂ data: Abgrall et al. (1993a) and Abgrall et al. (1993b). These are the data used by the McCandliss (2003) H2ools analysis package described above. The atomic data for interstellar Fe II lines (Howk et al. (2000)), interstellar Cl I lines (Sonnentrucker et al. (2006)), and stellar Fe II lines (Harper et al. (2001)) were also revised and should be considered for use in future studies.

8.4.2 FUSE Interstellar Absorption Spectra

Because a continuum source observed by FUSE with a color excess of only a few tenths will be littered with strong interstellar absorption lines, it is useful to have a rough idea of their expected strengths. Sembach (1999) provides lists of important interstellar features and detailed plots of synthetic interstellar spectra. The latter are especially useful for line identification purposes. The publications by McCandliss (2003) and Rachford et al. (2001) are also useful for their presentation of H₂ spectra.

8.4.3 Atlases of FUSE Stellar Spectra

Table 8.1 lists the atlases of FUV spectra that were produced during the operational lifetime of the FUSE satellite. These atlases provide a useful introduction to the FUV spectra of a wide variety of Galactic and extragalactic objects. In many cases, the processed spectra can be downloaded for immediate inspection. Consequently, these atlases are valuable resources for many purposes, especially for researchers who are new to the FUV region of the spectrum.

Table 8.1: Published Atlases of FUSE Spectra

Spectra	Reference	Online Availability
AG Dra (Symbiotic Star)	Young [2005]	…
Cool Stars	Dupree [2005]	…
Galactic OB Stars	Pellerin [2002]	http://archive.stsci.edu/prepds/atlasfuse/
Magellanic OB Stars	Walborn [2002]	http://archive.stsci.edu/prepds/atlasfuse/
Wolf-Rayet Stars	Willis [2004]	http://archive.stsci.edu/prepds/fuse_wratlas/
Starburst Galaxies	Pellerin & Robert	http://archive.stsci.edu/prepds/fuse_galaxies/
Quasar Composite	Scott [2004]	http://archive.stsci.edu/prepds/composite_quasar/

8.4.4 FUSE Airglow Spectra

Because FUSE spectra are always contaminated to some level by airglow, it is useful to be familiar with airglow spectra. Feldman et al. (2001) present an extensive set of airglow spectra observed with FUSE.

8.4.5 IUE Object Classes and Class #99 Observations

With the availability of a large body of IUE data already in the MAST archive, the FUSE project adopted an only slightly modified version of the IUE Object classes for use by FUSE. Table 8.2 lists the FUSE Object Classes.

Under class 99, Nulls & Flat Fields, the FUSE project has entered the following data types:

STIMLAMP:

These are pseudo flat field exposures, taken with the internal lamps. They expose all of the flaws, grid wires and shadows on the FUSE detectors (see the Instrument Handbook 2009).

STIMPULSE:

These are exposures containing only the electronically injected pulses described in Chapter 2.

BKGND-DOORS-CLOSED:

These are essentially dark exposures. They were obtained with the outer doors of the FUSE telescopes closed at the end of the mission. The only events detected in these exposures are true dark counts and any stray light that may find its way onto the detectors through an unintended route.

MT-TEST:

This stands for moving target test. These observations were obtained as part of a program testing the ability of FUSE to follow a moving target.

Table 8.2: Object Classes

Class - Object	Class - Object	Class - Object
0-Sun	31-A4-A9 V-IV	60-Shell Star
1-Earth	32-A0-A3 III-I	61-ETA Carinae
2-Moon	33-A4-A9 III-I	62-Pulsar
3-Planet	34-AE	63-Nova-Like
4-Planetary Satellite	35-AM	64-Other
5-Minor Planet	36-AP	65-Misidentified Targets
6-Comet	37-WDA	66-Interacting Binaries
7-Interplanetary Medium	38-HB Stars	69-Herbig-Haro Objects
8-Giant Red Spot	39-Composite Sp Type	70-PN + Central Star
10-W C	40-F0-F2	71-PN - Central Star
11-WN	41-F3-F9	72-H II Region
12-Main Sequence O	42-FP	73-Reflection Nebula
13-Supergiant O	43-Late-Type Degen	74-Dk Cld (Abs Spectrum)
14-OE	44-G V-IV	75-Supernova Remnant
15-OF	45-G III-I	76 Ring Neb (Shk Ionized)
16-SD O	46-K V-IV	80-Spiral Galaxy
17-WD O	47-K III-I	81-Elliptical Galaxy
19-Other Strong Sources	48-M V-IV	82-Irregular Galaxy
20-B0-B2 V-IV	49-M III-I	83-Globular Cluster
21-B3-B5 V-IV	50-R, N, or S Types	84-Seyfert Galaxy
22-B6-B9.5 V-IV	51-Long Period Var Stars	85-Quasar
23-B0-B2 III-I	52-Irregular Variables	86-Radio Galaxy
24-B3-B5 III-I	53-Regular Variables	87-BL Lacertae Object
25-B6-B9.5 III-I	54-Dwarf Novae	88-Em Line Gal(Non-Seyf)
26-BE 27-BP	55-Classical Novae	90-Intergalactic Medium
28-SDB	56-Supernovae	98-Wavelength Cal Lamp
29-WDB	58-T Tauri	99-Nulls & Flat Fields
30-A0-A3 V-IV	59-X-Ray