Here we discuss the available MaNGA data products at MAST.

On this page...

Summary

The MaNGA survey produces four main data products from its output data reduction and analysis pipelines: 2-D row-stacked spectra and 3-D IFU flux data cubes output from the data reduction pipeline (DRP); 3-D model data cubes, and 2-D images of mapped properties (kinematics, emission-line fluxes, and spectral indices) from the data analysis pipeline (DAP)(kinematics, emission-line fluxes, and spectral indices).  Additionally, it produces two summary catalog files: the "drpall" and "dapall" files, one for each pipeline.  See MaNGA Data for a summary of the data products.  The figure below highlights representative data products, along with a image preview of a galaxy. 

The full list of products can be viewed in the "Available SDSS Data Products at MAST" table on the Data Access at MAST page of this manual.

MaNGA Data

At MAST

The SDSS MaNGA observations are organized as one observation per galaxy, with the name "sdss_manga_[PLATE]-[IFUDESIGN]", where PLATE is the SDSS plate id number, and IFUDESIGN is the IFU fiber bundle name, for example "sdss_manga_9514-12704"


At MAST, the SDSS MaNGA data products are organized by unique observation, with names of the format sdss_manga_[PLATE]-[IFUDESIGN], where PLATE is the SDSS plate id number, and IFUDESIGN is the IFU fiber bundle name, e.g. "sdss_manga_9514-12704".   Each observation contains approximately ~15 files, of the following types:

ProductTypeDescriptionMinimum Recommended Product (MRP)File Type
CUBEdatacube IFU data cubesYes - only LOGCUBEfits
RSSspectralrow-stacked spectraYes - only LOGRSSfits
MAPSmapsthe derived 2d analysis mapsYes - only HYB10-MILESHC-MASTARSSPfits
MODELCUBESdatacubeIFU cube of model spectral fitsYes - only HYB10-MILESHC-MASTARSSPfits
drpallcatalogsummary catalog of DRP outputs
fits
dapallcatalogsummary catalog of DAP outputs
fits
imageimagepreview gri image with IFU
png

Data Reduction Pipeline (DRP)

The primary data products of the DRP output are 2-D row-stacked spectra files, and 3-D IFU data cubes.   The filename syntax includes variables PLATE, the SDSS plate id number, and IFUDESIGN, the IFU fiber bundle name.

RSS files

  • manga-[PLATE]-[IFUDESIGN]-LINRSS.fits.gz
  • manga-[PLATE]-[IFUDESIGN]-LOGRSS.fits.gz
  • SDSS Data Model

These are the row-stacked, flux-calibrated fiber spectra for a given galaxy across all exposures. The "LOGRSS" file has logarithmic wavelength sampling from log10(lambda/Angstroms)=3.5589 to 4.0151 (NWAVE=4563 spectral elements). The "LINRSS" file has linear wavelength sampling from 3622.0 to 10353.0 Angstroms (NWAVE=6732 spectral elements). Both files contain one row for each fiber, for a total of NFIBER*NEXP rows.

In brief, these files contain extensions for the flux (in units of 10-17 erg/s/cm2/Angstrom/fiber), the inverse variance, the pixel mask, the pre- and post-pixellized spectral line spread function for each fiber, the wavelength vector, the median spectral resolution (pre- and post-pixellization) as a function of wavelength for the fibers in this IFU, the standard deviation of spectral resolution (pre- and post-pixellization) as a function of wavelength for the fibers in this IFU, a binary table describing the individual exposures that make up the file, and arrays of the effective X and Y positions in arcsec of each fiber (as a function of wavelength) relative to the IFU center.  See the RSS DataModel for a full description of the data structure.

Cube files

  • manga-[PLATE]-[IFUDESIGN]-LINCUBE.fits.gz
  • manga-[PLATE]-[IFUDESIGN]-LOGCUBE.fits.gz
  • SDSS Data Model

These are the final 3d data cubes for a given galaxy that combine all fiber spectra across all exposures as described here.

The LOGCUBE data cube has logarithmic wavelength sampling from log10(lambda/Angstroms)=3.5589 to 4.0151 (NWAVE=4563 spectral elements), and 0.5 arcsec spatial pixels (spaxels) for a total size of NX x NY x NWAVE pixels; LINCUBE is the same, except it has linear wavelength sampling from from 3622.0 to 10353.0 Angstroms (NWAVE=6732 spectral elements).

Similar to the RSS files, they contain extensions for the flux (now in units of 10-17 erg/s/cm2/Angstrom/spaxel), inverse variance, pixel mask, per-spaxel line spread function (pre- and post-pixellized versions), wavelength vector, median spectral resolution and standard deviation thereof (pre- and post-pixellized versions), and a binary table describing the individual exposures that make up the file. Additionally, they also include reconstructed broadband 'griz' images created from the spectral data cube, estimates of the reconstructed point source profiles in each of the 'griz' bands, and estimates of the 2d spatial correlation matrices at 'griz' bands.  See the Cube DataModel for a full description of the data structure.

Data Analysis Pipeline (DAP)

The primary data products of the DAP output are 2-D derived maps files and 3-D model data cubes.  The filename syntax includes variables PLATE, the SDSS plate id number, IFUDESIGN, the IFU fiber bundle name, and DAPTYPE, a short-hand signifying the method used to analyze the DRP LOGCUBE files.  

DAPTYPE

This is a combination of the keywords used to select the spaxel binning approach, the templates used by the stellar-continuum (stellar kinematics) fit, and the templates used during the continuum+emission-line modeling. DAPTYPE can have four values:

  • SPX-MILESHC-MASTARSSP: Analysis of each individual spaxel with S/N>1 ; spaxels must have a valid continuum fit for an emission-line model to be fit.
  • VOR10-MILESHC-MASTARSSP: Spaxels are binned to S/N~10 using the Voronoi binning algorithm; all binned spectra are treated independently.
  • HYB10-MILESHC-MASTARSSP: Stellar-continuum analysis of spectra binned to S/N∼10 for the stellar kinematics (same as VOR10 approach); however, the emission-line measurements are performed on the individual spaxels. Hierarchically clustered MILES templates are used for stellar continuum analysis and simple stellar population models derived from MaStar are used to fit the continuum in the emission-line module.
  • HYB10-MILESHC-MASTARHC2: Same as the above except hierarchically clustered templates from MaStar are used to fit the stellar continuum in the emission-line module.

The advantage of the VOR10-MILESHC-MASTARSSP output is that all measurements are performed on exactly the same spectra. The HYB10-MILESHC-MASTARSSP output is meant to allow for greater spatial resolution for the emission-line analysis and avoid limitations of that analysis from tying it to the S/N of the broadband continuum.

MAPS files

The MAPS files are the primary output file from the DAP and provide 2D "maps" (i.e., images) of DAP measured properties. The shape and WCS of these images identically match that of a single wavelength channel in the corresponding DRP LOGCUBE file. 

Most properties are provided in groups of three fits extensions:

  • [property]: the measurement value,
  • [property]_IVAR: the measurement uncertainty stored as the inverse variance
  • [property]_MASK: a corresponding bit mask for each pixel.

Extensions can either be a single 2D image or they can have a series of images that are organized along the third dimension. For the latter, each image is said to be in a specific "channel." For example, each Gaussian-fitted emission-line flux is provided in a single channel in the EMLINE_GFLUX extension. 

MODELCUBE files

The DAP model LOGCUBE files provide the binned spectra and the best-fitting model spectrum for each spectrum that was successfully fit. These files are useful for detailed assessments of the model parameters because they allow you to return to the spectra and compare the model against the data. The DAP fits the spectra in two stages, one to get the stellar kinematics and the second to determine the emission-line properties. The emission-line module (used for all binning schemes) fits both the stellar continuum and the emission lines at the same time, where the stellar kinematics are fixed by the first fit. The exact details of the stellar continuum fit determined by the stellar kinematics and emission-line modules are, therefore, different.

It's important to note that the DAP performs all spectral fitting on the binned spectra (termed as such even if a bin only contains a single spaxel) after they have been corrected for Galactic extinction. Therefore, the output emission-line fluxes have been corrected for Galactic extinction. However, the models and binned spectra in the output LOGCUBE file are reverted to their reddened values for direct comparison with the DRP LOGCUBE file.

Also, for the HYB binning case, the binned spectra provided in the LOGCUBE files are from the Voronoi binning step; however, the emission-line models are fit to the individual spaxels. Therefore, when using the LOGCUBE files for this binning scheme:

  •  The stellar-continuum fits (e.g., the data in the STELLAR extension) should be compared to the Voronoi binned spectra in the file, but
  •  the best-fitting model spectra (stellar continuum + gas emission) in the MODEL extension should be compared to the individual spectra from the DRP LOGCUBE file!

Catalog Files

There are two summary catalog files, one for each pipeline: the drpall file for the DRP, and the dapall, for the DAP.   The filename syntax for both files includes DRPVER, the version of the DRP used, and DAPVER, the version of the DAP used.  

drpall

 This is the final summary FITS binary table for the MaNGA Data Reduction Pipeline (DRP), and contains various metadata and header information taken from the output MaNGA data cubes. It contains no actual spectra from MaNGA cubes, but all of the information required to find a given set of spectra. It can be used to search for galaxies matching a given set of selection criteria.  See the drpall DataModel for a full description of the data structure. 

In particular, the DRPALL file provides two identifiers for MaNGA targets: MANGAID and PLATEIFU.

  •  PLATEIFU is composed of the plate ID (PLATE) and the IFU design ID (IFUDSGN) in the format of "PLATE-IFUDSGN". It uniquely identifies an observation of a target. The same object drilled on two different plates will have different PLATEIFU identifiers. 
  •  MANGAID is composed of a catalog ID number and an object number with a hyphen in between. Within the same input catalog, it uniquely identifies each object. The same galaxy from the same catalog always have the same MANGAID. 

dapall

The DAPall file provides final summary tables of the MaNGA Data Analysis Pipeline (DAP) results. It provides collated summary information about the DAP methods and global metrics derived from the DAP analysis products useful for sample selection. Similar to the DRPall file, the DAPall file contains no resolved maps or model spectra, but a set of data that can be useful for selecting subsets of MAPS and model LOGCUBE files for further analysis. See the dapall DataModel for a full description of the data structure. 

The DAP analyzes each observation (PLATEIFU) once per analysis method (DAPTYPE). However, not all datacubes provided by the DRP are analyzed by the DAP.  The DAPall file has multiple extensions, one for each DAPTYPE analysis. There is a row in each extension for all attempted analyses of the available PLATEIFU-DAPTYPE combinations. Some of the attempted analyses faulted, and these are listed among the known issues. The purpose of the DAPDONE column in the DAPall file is to select those PLATEIFU-DAPTYPE combinations where the DAP was successful in its analysis of the data.

For Further Reading...