Introduction and Instructions

[ Overview | How to use HIRES | Output ]


HIRES is a tool that allows users to produce higher spatial resolution images from archival data of the Infrared Astronomical Satellite (IRAS). HIRES uses the Maximum Correlation Method (MCM, H.H. Aumann, J.W. Fowler and M. Melnyk, 1990, AJ, 99, 1674).

HIRES can be a useful tool for studying morphology and for separating confused sources. It can produce spatial resolutions of better than an arcminute, roughly a five-fold increase over the unenhanced resolution. However, the actual resolution achieved for a given source varies in a complicated fashion.

HIRES results can be used to do aperture photometry. HIRES fluxes are accurate to about 20%. Most of the uncertainty is due to background estimation uncertainties.

HIRES images are available as 1 degree by 1 degree, or 2 degree by 2 degree, images, with 15" pixels.

The full IRAS documentation is here. However use the links on this page for the current HIRES information.

Note: If spatial resolution is not critical, the released IRAS Sky Survey Atlas (ISSA) images, in which all 4 bands are smoothed to the resolution of the 100 µm band, may be sufficient.

How to use HIRES:

The single-position web template requires a user-specified target ID and coordinates (or a source name resolvable by SIMBAD or NED). Output image size, background removal choice, and flux densities for the sample maps may be set, or the defaults can be used.

For table upload, give positions as J2000 equatorial coordinates under column names "ra" and "dec". The other parameters may be set globally, or within the table. A valid table entry will override the global value.

Suitable Targets

Turning on baseline removal does a better job of de-striping in most cases. As a linear baseline is fit to each scan, it also removes any DC background and/or planar gradient in the background. If baseline removal is turned off, HIRES will do cross-scan offset removal, which takes out a DC offset from each scan and tries to bring all scans to a common background. In either case, a second-pass destriper provides "fine" destriping; this is known as compensation offset destriping.

Because baseline removal results in many negative values, and the HIRES algorithm discards negatives, baseline removal can result in large areas of artificially low coverage. Cross-scan offset removal often leaves a high DC background, and point sources converge slowly and often "ring" on high background. For these reasons, a flux bias parameter (FBIAS) is automatically applied to add (for baseline removal), or subtract ( for cross-scan offset), a DC value to all the data BEFORE HIRES PROCESSING and subtract it out again afterwards. An optimum FBIAS value is calculated during the data preparation process that ensures the data are positive but as close to zero as possible during HIRES processing.

The IRAS "beam" is not Gaussian, nor is the resolution in the image the same across an entire image. HIRES produces beam sample maps ("bem" maps), which show how the achieved resolution depends on source strength and coverage. They consist of a field of point-source spikes superimposed on a highly smoothed version of the image (note: the image smoothing includes attenuation of point-source flux). The user can specify the flux densities of a point source of interest, in Jy, one value per band. The default is 10 Jy in all four bands. A regular grid of 5x5 samples for a 1x1, or 9x9 spikes for a 2x2, square-degree field is used.


The output comes in a tar file called "results.tar". A results page will come up at the end of processing, telling you how to retrieve it. The FITS images are in the sub-directory "img". The final images are the four 20-iteration files, one per IRAS band, labeled "img_*_20.fits".

HIRES images are on the AC scale (instead of the DC scale). The standard point-source correction factors have already been applied to the data. A correction back to the DC scale may be necessary for very extended structure (Page IV-9 of the IRAS Explanatory Supplement).

NOTE: The output images are in a cylindrical rectangular (-CAR) projection. But the headers indicate a standard gnomonic (-TAN) projection. This is done to avoid problems with FITS readers unfamiliar with the -CAR projection. For an individual image of the sizes generated here, the differences are minor. Also, the output image headers are in B1950 coordinates.

HIRES Output Files

HIRES Image Types

Interpreting HIRES Images

Artifacts in HIRES Images

IRAS Data Characteristics