This tutorial runs through all of the steps for doing a project with
Marvin from start-to-finish with no extra fat. We recommend the use of
ipython or jupyter notebook when using Marvin. You can start either
from a terminal with ipython or jupyter notebook.
Calculate the [NII]/H$\alpha$ ratio for star-forming spaxels in galaxies with stellar mass between $10^{10}$ and $10^{11}$ .
Marvin uses a simplified query syntax (in both Web and local queries) that understands the MaNGA database schema, so you don't have to write complicated SQL queries.
Goal: Find galaxies with stellar mass between $10^{10}$ and $10^{11}$.
Create the query with ~marvin.tools.query.query.doQuery and run it
(limit to only 3 results for demo purposes):
In [ ]:
from marvin.tools.query import doQuery
q, r = doQuery(search_filter='nsa.sersic_logmass >= 10 and nsa.sersic_logmass <= 11', limit=3)
Tip: see Marvin Query to learn the basics of querying. See Example Queries and Marvin Query Syntax Tutorial for help with designing search filters.
View the ~marvin.tools.query.results.Results. You may see a different
set of results. That is ok as long as you see some set of results.:
In [ ]:
df = r.toDF()
df
Convert into ../tools/maps objects:
In [ ]:
r.convertToTool('maps')
r.objects
galaxies = r.objects
In [ ]:
from marvin.tools.maps import Maps
mangaids = ['1-245458', '1-22301', '1-605884']
galaxies = [Maps(mangaid=mangaid) for mangaid in mangaids]
Get the H$\alpha$ maps:
In [ ]:
haflux_maps = [galaxy['emline_gflux_ha_6564'] for galaxy in galaxies]
Plot H$\alpha$ map of the second galaxy:
In [ ]:
haflux_map = haflux_maps[1]
fig, ax = haflux_map.plot()
Let's take a look at the model fits a spaxel. The easiest way is to navigate to the Galaxy page for 7992-6101 and click on the red "Map/SpecView Off" button.
However, we can also plot the spectrum and model fits in Python. First, we can find the coordinates of a spaxel by moving our cursor around the interactive matplotlib plotting window. When the cursor is over the spaxel of interest, the coordinates will appear in the lower right.
Then we can create a ~marvin.tools.spaxel.Spaxel object by accessing
the parent ~marvin.tools.maps.Maps object from the
~marvin.tools.quantities.Map object (haflux_map.maps) and retrieve
the model fit.
In [ ]:
spax = galaxies[1].getSpaxel(x=28, y=24, xyorig='lower', cube=True, modelcube=True)
Now let's plot the spectrum and model fit:
In [ ]:
import matplotlib.pyplot as plt
# Set matplotlib style sheet. Undo with matplotib.rcdefaults().
plt.style.use('seaborn-darkgrid')
In [ ]:
ax = spax.flux.plot()
ax.plot(spax.full_fit.wavelength, spax.full_fit.value)
ax.legend(list(ax.get_lines()), ['observed', 'model'])
ax.axis([7100, 7500, 0.3, 0.65])
In [ ]:
masks, fig, axes = galaxies[1].get_bpt()
For a detailed description see BPT Diagrams.
Select the star-forming spaxels that are in the star-forming region of each diagnostic diagram (hence the "global" keyword):
In [ ]:
sf = masks['sf']['global']
Return the complement of the BPT global star-forming mask (True means
star-forming) using ~ and mark those spaxels as DONOTUSE since they
are non-star-forming spaxels.
In [ ]:
mask_non_sf = ~sf * haflux_map.pixmask.labels_to_value('DONOTUSE')
Do a bitwise OR between the DAP mask and the non-star-forming mask:
In [ ]:
mask = haflux_map.mask | mask_non_sf
Plot with our new mask:
In [ ]:
haflux_map.plot(mask=mask)
In [ ]:
maps_7992_6101 = galaxies[1]
nii = maps_7992_6101['emline_gflux_nii_6585']
ha = maps_7992_6101['emline_gflux_ha_6564']
nii_ha = nii / ha
Plot the [NII]/H$\alpha$ flux ratio for the star-forming spaxels:
In [ ]:
nii_ha.plot(mask=mask, cblabel='[NII]6585 / Halpha flux ratio')
In [ ]: