Release: Iris 2.0 – SED analysis tool

by Guest on September 18, 2013

This post is written by Jamie A. Budynkiewicz who works with the Chandra Software group at the Smithsonian Astrophysical Observatory.

Spectral energy distributions (SEDs) provide a wealth of knowledge about galaxies: their type, nuclear activity, luminosity, mass, star formation rate – basically, SEDs are pretty darn useful. Analyzing an SED involves loading the data, plotting it, inspecting it, and finally fitting it with models to derive meaningful physical quantities. However, we can easily spend most of our time writing a program to do this than actually analyzing the physics from the SED.

Enter Iris, the Virtual Astronomical Observatory’s broadband SED analysis tool.

Iris delivers an easy-to-use interface for building, visualizing and fitting SEDs. Iris seamlessly combines SpecView’s GUI and data manipulation features, Sherpa’s robust modeling and fitting capabilities, and streamlined access to NED’s extragalactic SED database (as well as data from file) into one desktop application.

The VAO recently released Iris 2.0, the latest version of the SED analysis tool. Iris 2.0 comes with new powerful science capabilities which allow the redshifting, interpolation and smoothing of SEDs, and measurement of integrated fluxes through simple spectral bandpasses or through one of the thousands of photometric filters provided by the Spanish VO Filter Profile Service. Users can also apply simple aperture corrections to individual photometric points or to whole SED segments. In addition to improved visualization capabilities, including the co-plotting of SEDs, Iris 2.0 has increased interoperability with other VO tools.

Getting Started with IRIS

Download Iris 2.0 — Iris is available for 32- and 64-bit Linux and Mac OS X 10.5, 10.6, 10.7, 10.8 systems with Java v1.6 or above (special download instructions exist for Mac OS X 10.5 and Java 1.5).

 For those of you unfamiliar with Iris, check out this tutorial video to catch a glimpse at some of Iris’ capabilities. I’ve described a few of my favorite features below.

The SED data handling is straightforward. You can import your own SED data in FITS, VOTable, ASCII, CSV, TST or IPAC format, or load data from NASA/IPAC Extragalactic Database or the Italian Space Agency Science Data Center from the Iris desktop [Iris recognizes IVOA compliant metadata in FITS or VOTable format, which includes bibliographic information, the spectral bandpasses used for the data, comments about the data reduction, and other reference information – useful for close data inspection]. You can build the SED by adding/removing multiple, separate data segments or photometric points. Many separate SEDs can be loaded into Iris in a given session as well. When SEDs are uploaded, the data is automatically plotted in the display window; SEDs with multiple segments are color-coded by segment.

Iris’ visualization capabilities give you easy control over how/what data is displayed. If any of you are like me and find unit conversions tedious and error-prone, you’ll love Iris’ plot window. Here, Iris lets you easily change the flux and spectral axes units into a variety of useful astronomical units and lets you switch the Y-axis between flux density and flux. You can also co-plot multiple, separate SEDs.

Iris allows smooth access to SED data throughout the analysis. Simply right-clicking on a data point displays its metadata, spectral coordinates, uncertainties and segment information, and allows you to apply a simple aperture correction to that point or to the whole SED segment the point belongs to. The Metadata Browser is a powerful inspection tool, allowing you to view every point’s data and metadata, hierarchically sort data each column, filter data using a Boolean expression, and highlight selected regions from the plot window. You can also send selected data back and forth between Iris and other Virtual Observatory tools, like Topcat and VOPlot, via SAMP (Simple Application Messaging Protocol), which is useful if you want to plot something other than an SED.

Iris provides an interactive fitting tool for modeling SEDs. While it’s not-so-great for those of you wanting to fit hundreds of SEDs simultaneously, the Iris fitting tool lets you carefully fit one SED at a time. You can import your own custom templates, template libraries and Python functions to Iris; these models/templates are saved in Iris for future sessions. Iris also comes with a long list of astrophysical and mathematical models, like blackbody, log parabola, bremsstrahlung radiation, dust extinction curves, and Gaussian absorption/emission features (Iris adopts its models from Sherpa). You can build a user-defined model from a combination of preset/custom models and templates, and save the model for a later session. Iris offers robust optimization methods and fit statistics, with suggestions in the User’s Guide on choosing the best statistics for your data. Confidence limits on model parameters can also be calculated.

Lastly, download and installation is painless. Simply download the tar file for your computer platform, unpack it, and that’s it. You’re all set to analyze SEDs in Iris!

For more information, please visit the Iris online manual at http://cxc.harvard.edu/iris.

Iris is an active project. We are regularly adding new requirements for future releases, such as improved template library handling and bulk-downloading of SEDs. If you have any comments or suggestions, feel free to tell us here or send a message at http://cxc.harvard.edu/helpdesk/!

{ 5 comments… read them below or add one }

1 Rick Ebert October 2, 2013 at 1:11 pm

The link to the NED VO service page is incorrect: it should be: http://vo.ned.ipac.caltech.edu/
The user interface for access to NED SEDs may be accessed from the main NED webpage.

Reply

2 Jamie Budynkiewicz October 8, 2013 at 3:16 pm

Oops — Thanks for posting the correct link, Rick.

3 Joseph Wang October 16, 2013 at 7:11 pm

Is the source code for Iris downloadable. I’d like to repackage IRIS as a rpm so that it can be included in the next distributions of Mageia and Fedora linux.

Reply

4 Jamie Budynkiewicz October 17, 2013 at 3:45 pm

Hi Joseph,

Thank you for your interest in Iris!

Yes – you can download the source code from the VAO repository following these directions. And just to let you know, we will be releasing a patch for Iris 2.0 around the end of October.

If you need any help or would like more information on Iris, feel free to contact me directly at jbudynkiewicz at cfa.harvard.edu.

5 Joseph Wang October 18, 2013 at 5:10 am

Thanks. I’ll see to it that IRIS gets into the next version of Mageia and the RPM package should be something that will also get it into Fedora.

Reply

Leave a Comment

Previous post:

Next post: