Skip to article frontmatterSkip to article content
SPHEREx Archive at IRSA

SPHEREx Data Products

A detailed description of SPHEREx data products available to the public is provided in the SPHEREx Explanatory Supplement. Here we provide a concise summary of the science, calibration, and additional data products available at IRSA. This summary includes filenaming conventions, for which we adopt the following definitions:

Main Science Data Product: Spectral Image Multi-Extension FITS Files (MEF)

The main Quick Release data product is the Level 2 Spectral Image MEF,as described in Section 2.1 of the Explanatory Supplement.

There are 6 Spectral MEFs (one for each detector) for each sky pointing. Because data quality assessments are evaluated per spectral image band, some observations will not include all 6 bands in the archive. Each Spectral MEF is approximately 70 MB and contains 6 extensions:

HDU 1: IMAGE

Calibrated surface brightness flux density in units of MJy/sr, stored as a 2040 x 2040 image. No zodiacal light subtraction is applied.

The SPHEREx focal plane is split with a dichroic to three short-wavelength and three long-wavelength detector arrays. Two focal plane assemblies (FPAs) simultaneously image the sky through a dichroic beam splitter. Each FPA contains three 2K x 2K detector arrays placed behind a set of linear variable filters (LVFs), providing narrow-band response with a band center that varies along one axis of the array. SPHEREx obtains spectra through multiple exposures, placing a given source at multiple positions in the field of view, where it is measured at multiple wavelengths by repointing the spacecraft.

  • Band 1: λ= 0.75 - 1.09 µm; R=39
  • Band 2: λ= 1.10 - 1.62 µm; R=41
  • Band 3: λ= 1.63 - 2.41 µm; R=41
  • Band 4: λ= 2.42 - 3.82 µm; R=35
  • Band 5: λ= 3.83 - 4.41 µm; R=112
  • Band 6: λ= 4.42 - 5.00 µm; R=128
HDU 2: FLAGS

Bitmap of per-pixel status and processing flags, stored as a 2040 x 2040 image. The definition of the flags are provided in Table 10 of the SPHEREx Explanatory Supplement.

HDU 3: VARIANCE

Variance of calibrated surface brightness flux in units of (MJy/sr)^2, stored as a 2,040 x 2,040 image.

HDU 4: ZODI

Modeled zodiacal light background flux in units of MJy/sr, stored as a 2040 x 2040 image. This has not been subtracted from the IMAGE extension.

HDU 5: PSF

121 Point-spread functions (PSFs); each PSF is represented as a 101 x 101 image and all 121 are assembled together into a cube.

HDU 6: WCS-WAVE

Spectral World Coordinate System (WCS) FITS-compliant lookup table that maps spectral image pixel coordinates to central wavelengths and bandwidths. The lookup table consists of 1 row with 3 columns (X, Y, VALUES).

X and Y are each arrays defining a grid of control points in spectral image pixel space.

VALUES is an array of two-element arrays: at each (X, Y) control point, the two-element array contains the central wavelength and the corresponding bandwidth.

Filename Format:

Example:

Cutouts of Spectral Image MEFs

IRSA enables users to access rectangular cutouts of a SPHEREx Spectral Image MEF by simply appending a query string containing center and size parameters to the image URL. These cutout MEFs contain the same HDUs as the original Spectral Images (IMAGE, FLAGS, VARIANCE, ZODI, PSF, WCS-WAVE). However, the IMAGE, FLAGS, VARIANCE, AND ZODI HDUs have been modified to include only those pixels within the specified cutout size. The WCS-WAVE HDU has also modified to provide the correct mapping between the pixels in the cutout to wavelength. The PSF HDU from the original spectral image is included unmodified in the cutout MEF.

The spatially-varying PSF is represented as an image cube with 121 planes. Each plane is a 101x101 pixel image representing a PSF for a different region of the detector. Users interested in performing photometry on a cutout using the information in the cutout PSF HDU will need to understand how to find the most applicable PSF cube plane for each pixel in the cutout. The basic steps are described below, and a Python notebook tutorial is provided to help users get started with a simple implementation.

  1. Determine the 0-based pixel coordinates of the position of interest in the IMAGE HDU of the cutout.

  2. Determine the 0-based pixel coordinates of the position of interest in the IMAGE HDU of the original Spectral Image.

xpix_orig = 1 + xpix_cutout - CRPIX1A
ypix_orig = 1 + ypix_cutout - CRPIX2A
  1. Examine the header of the PSF HDU of the cutout to determine the PSF zone and cube plane corresponding to the pixel of interest in the original Spectral Image.

The PSF HDU has a header containing the keywords XCTR_*, YCTR_*, XWID_*, and YWID_*, where * goes from [1 to 121]. To determine if a pixel in the original Spectral Image falls within a PSF zone, simply find the closest XCTR_* and YCTR_* to determine the cube plane that contains the corresponding PSF for this zone.

Calibration Product: Absolute Gain Matrix

The Absolute Gain Matrix products are ~16 MB FITS image files (one per detector) with dimensions 2,040 × 2,040 and units of (MJy/sr) / (e−/s).

Filename Format:

Example:

Calibration Product: Exposure-Averaged Point Spread Functions (PSFs)

The Exposure-Averaged Point Spread Functions (PSFs) are ~6 MB FITS cubes (one for each detector) with dimensions 101 × 101 × 121. Each of the 121 layers represents a “super-resolution” PSF estimate in a different region (defined by an 11x11 grid) of the detector. Each PSF is a two-dimensional array with size of 101 × 101 pixels. The PSFs are oversampled such that 10 PSF pixels cover the same spatial extent as one spectral image pixel (0.615 arcsec).

Filename Format:

Example:

Calibration Product: Dark Current

The Dark Current products are ~16 MB FITS image files (one per detector) with dimensions 2,040 x 2,040 and units of electron/s.

Filename Format:

Example:

Calibration Product: Dichroic

The Dichroic products are ~16 MB FITS image files with dimensions 2,040 x 2,040. The pixel value is 0 for pixels that are unaffected by flux attenuation due to the dicroic filter and 1 for impacted pixels. A pixel is considered impacted if the flux attenuation is 50% or higher. Note that only bands 3 and 4 have any non-zero values.

Filename Format:

Example:

Calibration Product: Electronic Gain Factors

The Electronic Gain Factor product is a single YAML file that includes the provenance information for the detectors and a list of 32 gain values per detector.

Filename Format:

Example:

Additional Product: Nonfunctional Pixels

The Nonfunctional Pixel products are ~32 MB FITS image files (one per detector) with dimensions 2,040 x 2,040. Pixel values are 1 for pixels known to be permanently non-functioning and 0 otherwise.

Filename Format:

Example:

Additional Product: Nonlinearity Parameters

The Nonlinearity Parameter products are ~79 MB multi-extension FITS files (one per detector). Each file contains 5 extensions (Q_nl, b1, b2, b3, Qmax), each of which is an image with dimensions 2,040 x 2,040. These extensions are described in Section 3.2.1 of the SPHEREx Explanatory Supplement.

Filename Format:

Example:

Additional Product: Read Noise Parameters

The Read Noise Parameters products are ~48 MB multi-extension FITS files (one per detector). Each file has 2 extensions: READNOISE-1 and READNOISE-2. Each extension is an image with dimensions 2040 × 2040 and units of electrons.

Filename Format:

Example:

Additional Product: Spectral WCS

The Spectral WCS products are ~32 MB multi-extension FITS files (one per detector). Each file has 3 extensions: CWAVE, CBAND, and WCS-WAVE. For science analysis, use the values in the CWAVE and CBAND layers, not the WCS-WAVE which is intended for visualization.

CWAVE is an image with dimensions 2,040 x 2,040. It contains the central wavelength in microns for each pixel.

CBAND is an image with dimensions 2,040 x 2,040. It contains the bandwidth in microns for each pixel.

WCS-WAVE is a table with 3 columns (X, Y, VALUES) and 1 row. This is equivalent to the WCS-WAVE extension in the Spectral Image MEF file described above.

Filename Format:

Example:

SPHEREx Archive at IRSA
SPHEREx Data Access