Maps and Map objects form a hierarchy:

  • Maps: set of DAP maps for a galaxy (analogous to a DAP MAPS FITS file),

  • Map: an individual map, and

  • EnhancedMap: an individual map modified by map arithmetic.

Maps is a class to interact with the set of DAP maps for a galaxy. For a general introduction to Marvin Tools, check out the Galaxy Tools section. Here we will revisit those features and will expand on some specifics of the Maps class.

Initializing a Maps

A Maps can be initialized by filename, plateifu, or mangaid.


>>> maps = Maps(filename='/Users/username/manga/spectro/analysis/v2_4_3/2.2.1/HYB10-GAU-MILESHC/8485/1901/manga-8485-1901-MAPS-HYB10-GAU-MILESHC.fits.gz')
>>> maps
<Marvin Maps (plateifu='8485-1901', mode='local', data_origin='file', bintype='HYB10', template='GAU-MILESHC')>

Either the full path or the path relative to the current working directory is required. A Maps initialized from a file will always be in local mode.

Plateifu or Mangaid:

>>> maps = Maps(plateifu='8485-1901')
>>> maps
<Marvin Maps (plateifu='8485-1901', mode='local', data_origin='db', bintype='HYB10', template='GAU-MILESHC')>

>>> maps = Maps(mangaid='1-209232')
>>> maps
<Marvin Maps (plateifu='8485-1901', mode='local', data_origin='db', bintype='HYB10', template='GAU-MILESHC')>

Marvin first attempts to find the data in a local database, otherwise will retrieve the data in remote mode.

Smart Galaxy Lookup

You can also initialize a Maps without the filename or a galaxy identifier (plateifu/mangaid) keyword argument, and Marvin will attempt to parse the input and find the desired galaxy:

>>> maps = Maps('8485-1901')
>>> maps
<Marvin Maps (plateifu='8485-1901', mode='local', data_origin='db', bintype='HYB10', template='GAU-MILESHC')>


The default Maps bintype is HYB10, where the stellar continuum analysis of spectra is Voronoi binned to S/N~10 for the stellar kinematics; however, the emission line measurements are performed on the individual spaxels. You can specify a different binning scheme with the bintype keyword (currently, the only other option is VOR10, which does the stellar continuum and emission line analyses on spectra Voronoi binned to S/N~10):

>>> maps = Maps('8485-1901', bintype='HYB10')
>>> maps
<Marvin Maps (plateifu='8485-1901', mode='local', data_origin='db', bintype='HYB10', template='GAU-MILESHC')>


Currently, the only template available is GAU-MILESHC, which is selected by default.

Basic Attributes

Like Cubes, Maps come with some basic attributes attached (e.g., the full header, the WCS info, the bintype and template) plus the NSA and DAPall catalog parameters.

# access the header
>>> maps.header

# access the wcs
>>> maps.wcs

# the NSA catalog information
>>> maps.nsa['z']

# the DAPall catalog info
>>> maps.dapall['sfr_tot']

Maps also has the DAP data quality, targeting, and pixel masks available as the quality_flag, target_flags, and pixmask attributes, respectively. These are represented as Maskbit objects.

Maps DataModel

The DAP datamodel is attached to Maps as the datamodel attribute. The datamodel describes the contents of the MaNGA DAP Maps, for a given release, and contains a list of Properties associated with a Maps. This is a subset of the full DAP datamodel only pertaining to Maps.

# display the datamodel for maps properties
>>> maps.datamodel
[<Property 'spx_skycoo', channel='on_sky_x', release='2.1.3', unit=u'arcsec'>,
 <Property 'spx_skycoo', channel='on_sky_y', release='2.1.3', unit=u'arcsec'>,
 <Property 'spx_ellcoo', channel='elliptical_radius', release='2.1.3', unit=u'arcsec'>,
 <Property 'spx_ellcoo', channel='elliptical_azimuth', release='2.1.3', unit=u'deg'>,
 <Property 'spx_mflux', channel='None', release='2.1.3', unit=u'1e-17 erg / (cm2 s spaxel)'>,
 <Property 'spx_snr', channel='None', release='2.1.3', unit=u''>,
 <Property 'binid', channel='binned_spectra', release='2.1.3', unit=u''>,
 <Property 'binid', channel='stellar_continua', release='2.1.3', unit=u''>,
 <Property 'binid', channel='em_line_moments', release='2.1.3', unit=u''>,
 <Property 'binid', channel='em_line_models', release='2.1.3', unit=u''>,
 <Property 'binid', channel='spectral_indices', release='2.1.3', unit=u''>,
 <Property 'specindex_corr', channel='tio2sdss', release='2.1.3', unit=u'Angstrom'>,
 <Property 'specindex_corr', channel='d4000', release='2.1.3', unit=u''>,
 <Property 'specindex_corr', channel='dn4000', release='2.1.3', unit=u''>,
 <Property 'specindex_corr', channel='tiocvd', release='2.1.3', unit=u''>]

Each Property in the datamodel describes an available Map inside the Maps container, and has a channel, units, and a description. You can fuzzy search through the list to identify maps:

# find the H-alpha Gaussian flux property
>>> maps.datamodel['gflux_ha']
<Property 'emline_gflux', channel='ha_6564', release='2.1.3', unit=u'1e-17 erg / (cm2 s spaxel)'>

Accessing an Individual Map

The Property`s provide an interface to extract and create an individual `Map. You can select an individual Map in one of four ways:

  • exact key slicing,

  • dot syntax,

  • getMap method, or

  • fuzzy key slicing.

>>> from import Maps
>>> maps = Maps(plateifu='8485-1901')

# exact key slicing
>>> ha = maps['emline_gflux_ha_6564']

# dot syntax
>>> ha = maps.emline_gflux_ha_6564

# getMap()
>>> ha = maps.getMap('emline_gflux_ha_6564')
# equivalently
>>> ha = maps.getMap('emline_gflux', channel='ha_6564')

# fuzzy key slicing
>>> ha = maps['gflux ha']

Fuzzy key slicing works if the input is unambiguously associated with a particular key:

# Unambiguous inputs
>>> maps['gflux ha']        # == maps['emline_gflux_ha_6564']
>>> maps['gvel oiii 5008']  # == maps[emline_gvel_oiii_5008]
>>> maps['stellar sig']     # == maps['stellar_sigma']

# Ambiguous inputs
# There are several velocity properties (stellar and emission lines).
>>> maps['vel']  # ValueError

# There are two [O III] lines.
>>> maps['gflux oiii']  # ValueError


Fuzzy key slicing is textual, not contextual, so sometimes inputs will be matched to unexpected keys. Fuzzy string matching can be fooled!

>>> maps.getMap('emline_gflux', 'H-beta')         # target channel: Hb_4862
<Marvin Map (property='emline_gflux_heta_3836')>  # actual channel: Heta_3836

In this case, the input 'H-beta' is supplied by the user with the understanding that the Balmer beta line at 4862 Angstroms is being requested. Unfortunately, that input matches more closely to they key 'Heta_3836' than it does to 'Hb-4862', but the match to 'Hb-4862' is not strong enough for the input to be labeled as ambiguous and raise an exception. When developing analysis scripts, check your inputs carefully, and make sure that data retrieval making use of fuzzy string-matching retrieves the correct channels!

Accessing an Individual Spaxel

Slicing a Maps returns a Spaxel object with all of its properties:

>>> sp = maps[9, 10]
>>> print(sp)
<Marvin Spaxel (plateifu=8485-1901, x=10, y=9; x_cen=-7, y_cen=-8, loaded=maps)>

Getting Bin IDs

For binned Maps, you can retrieve a Map of the bin IDs directly from the binid_* attributes. There are five types of bin IDs, designated as binid_[name]. You can list them from the datamodel:

>>> maps.datamodel.parent['binid']
<MultiChannelProperty 'binid', release='2.2.1', channels=['binned_spectra', 'stellar_continua', 'em_line_moments', 'em_line_models', 'spectral_indices']>

They are available as attributes.

# get a Map of the binned_spectra binids
>>> maps.binid_binned_spectra
<Marvin Map (property='binid_binned_spectra')>
[[-1. -1. -1. ..., -1. -1. -1.]
 [-1. -1. -1. ..., -1. -1. -1.]
 [-1. -1. -1. ..., -1. -1. -1.]
 [-1. -1. -1. ..., -1. -1. -1.]
 [-1. -1. -1. ..., -1. -1. -1.]
 [-1. -1. -1. ..., -1. -1. -1.]]

You can also retrieve a 2-d array of the bin IDs using the get_binid method. By default, get_binid will return the bin IDs for the binned_spectra channel of BINID.

# get the default binids
>>> maps.get_binid()
<Marvin Map (property='binid_binned_spectra')>
[[-1. -1. -1. ... -1. -1. -1.]
 [-1. -1. -1. ... -1. -1. -1.]
 [-1. -1. -1. ... -1. -1. -1.]
 [-1. -1. -1. ... -1. -1. -1.]
 [-1. -1. -1. ... -1. -1. -1.]
 [-1. -1. -1. ... -1. -1. -1.]]

# equivalent
>>> stvel_binids = maps.get_binid(property=maps.datamodel.stellar_vel)

Accessing Other Marvin Objects for the Same Galaxy

You can grab the associated DRP Cube with getCube:

>>> maps.getCube()
<Marvin Cube (plateifu='8485-1901', mode='remote', data_origin='api')>

or the Modelcube object using the getModelCube method:

>>> maps.getModelCube()
<Marvin ModelCube (plateifu='8485-1901', mode='remote', data_origin='api', bintype='HYB10', template='GAU-MILESHC')>

Saving and Restoring

You can save a Maps locally as a Python pickle object, using the save method:


Your saved Maps can be restored as a Maps object using the restore class method:

>>> from import Maps
>>> maps = Maps.restore('mymaps.mpf')

BPT Diagram

You can create a BPT diagram:

>>> masks, fig, axes = maps.get_bpt()


Class Inheritance Diagram

Inheritance diagram of

Class[input, filename, …])

A class that represents a DAP MAPS file.


Returns the binid map associated with a property.

Returns a version of self corresponding to the unbinned Maps.[method, …])

Returns the BPT diagram for this target.

Returns the Cube for with this Maps.

Returns the ModelCube for with this Maps.[x, y, ra, …])

Returns the Spaxel matching certain coordinates.

Retrieves a Map object.…)

Returns a ratio Map.

Downloads the maps using sdss_access - Rsync[path, overwrite])

Pickles the object.[, delete])

Restores a MarvinToolsClass object from a pickled file.