Measuring 5-sigma Depth Curves

In this notebook we will extract an object light curve from the Twinkles field, and measure the 5-sigma limiting depth at each epoch. The reason to do this is to start trying to understand the error properties of the Monitor light curves (including their biases) as a function of observation properties, such as image quality and image depth.

Requirements

You will need the DESC Monitor and its dependencies.

You will also need to set up an SSH tunnel to NERSC scidb where the Twinkles PServ data is stored. Follow directions here and use the following code from the command line.

ssh -L 3307:scidb1.nersc.gov:3306 $USER@cori.nersc.gov



In [1]:

    
import desc.monitor
import matplotlib.pyplot as plt
%matplotlib inline
%load_ext autoreload
%autoreload 2









    



/global/common/cori/contrib/lsst/lsstDM/twinkles_20170208/Linux64/sims_catalogs/2.3.3.3.sims+5/python/lsst/sims/catalogs/db/dbConnection.py:470: UserWarning: Duplicate object type id 25 specified: 
Output object ids may not be unique.
This may not be a problem if you do not want globally unique id values
  'want globally unique id values')
/global/common/cori/contrib/lsst/lsstDM/twinkles_20170208/Linux64/sims_catalogs/2.3.3.3.sims+5/python/lsst/sims/catalogs/db/dbConnection.py:470: UserWarning: Duplicate object type id 40 specified: 
Output object ids may not be unique.
This may not be a problem if you do not want globally unique id values
  'want globally unique id values')
/global/common/cori/contrib/lsst/lsstDM/twinkles_20170208/Linux64/sims_catalogs/2.3.3.3.sims+5/python/lsst/sims/catalogs/db/dbConnection.py:470: UserWarning: Duplicate object type id 4 specified: 
Output object ids may not be unique.
This may not be a problem if you do not want globally unique id values
  'want globally unique id values')

An Example Object Light Curve

Let's pull out one of the Twinkles objects and visualize it.



In [2]:

    
dbConn = desc.monitor.DBInterface(database='DESC_Twinkles_Level_2',
                                  #if running from ssh-tunnel uncomment below
                                  #host='127.0.0.1', port=3307,
                                  #or if running jupyter-dev uncomment below
                                  host='scidb1.nersc.gov', port=3306,
                                  driver='mysql')



In [3]:

    
lc = desc.monitor.LightCurve(dbConn)
lc.build_lightcurve_from_db(objid=48253)









    



WARNING: ErfaWarning: ERFA function "dtf2d" yielded 1120 of "dubious year (Note 6)" [astropy._erfa.core]



In [4]:

    
fig = lc.visualize_lightcurve()

The Matching Depth Curve

Now let's measure the 5-sigma limiting depth (for a point source). We do this by selecting a number of stars from the field, and then for each epoch, querying their flux errors, converting to limiting depth, and then averaging (with sigma-clipping) over the ensemble. All this is done by the measure_depth_curve() method.



In [5]:

    
worker = desc.monitor.Monitor(dbConn)
worker.get_lightcurves([48253])



In [6]:

    
worker.return_lightcurve[48253].visualize_lightcurve(using='flux')
plt.show()

We have three methods for calculating visit depth: DM, DM_modified, or stars. DM and DM_modified take the sky_noise and seeing values from the visit table in the NERSC database to do the calculations. stars uses the flux errors in the stars observed in the Twinkles visits to make the calculation. More information is available in our error model notebook also in this folder.



In [7]:

    
dc = worker.measure_depth_curve(using='DM_modified')



In [8]:

    
fig = worker.return_lightcurve[48253].visualize_lightcurve(using='mag')
dc.visualize_lightcurve(using='mag', include_errors=False, use_existing_fig=fig)
plt.show()



In [9]:

    
sc = worker.measure_seeing_curve()



In [10]:

    
fig = sc.visualize_seeing_curve()
plt.show()

Conclusion

We show here the monitor's basic lightcurve functionality and its ability to connect to a science database at NERSC to pull out relevant visit information and build lightcurves for objects. This is a valuable tool that makes it easy for anybody to do their own Twinkles analysis.