LSST Tutorials

This page consists of tutorials for people getting started with the LSST project and software.

First you need to get a physics department computer account.  Use your physics department account and login (using  ‘ssh -Y’) to

If you don’t know how to use unix at all then I suggest you start by looking at the book “Effective Computation in Physics”.

There is a copy in the student office and will give you instructions on how to get started.  Also, ask other students for help.  Make sure you can login, list files in your directory, edit a file etc.  When you to do basic operations you are ready to get started.  Alternatively, you can find many resources on line to introduce you to Unix.

You could also look at the beginning of the Super-K tutorials if you don’t know about unix (But be a bit careful if you look at this.  That uses the ‘tcsh’ and we use ‘bash’ for the shell)

Step 0: Computer Setup

To start, you will need a Duke physics department computer account. Talk to someone if you don’t.  Then, login to and type:
source /var/phy/project/lsst/

and this will setup the the LSST software for you.  First we will learn a bit about the program “ds9”.  It is an astronomical display program.  After doing this setup you can run it by just typing  ‘ds9’

You can play with ds9 a bit using this tutorial:

BTW that astrobites site is astronomy for undergrads written by graduate students.  There is a lot of good articles there; start reading it! Or consider writing for it!
To start, do the tutorial and make a  presentation explaining what you did for Friday group meeting. Try to follow the instructions in the ds9 tutorial and show us the steps to get to the last plots in your presentation. The links to the files (the FITS) files no longer work.  But, you can go here to get them:
Then, here is a tutorial you can play with on FITS files.  This will use ipython to look at data in an image.  You should know how to start a iPython or Jupyter notebook from your physics classes already.  If not ask one of the other LSST people.
Go ahead and do this.  Some of the other astropy ones are good too. Play with these to get used to using the tools.  Next you might want try doing something a bit more complex.  Our LSST colleague Robert Lupton developed an algorithm for optimally combining images from multiple color bands together.  That algorithm is implemented in astropy here:
You could see if you could use this on your image above.
None of this work is specific to LSST.  One thing people here have been working on is a program called GalSim.
This is a program to make simulated images of galaxies and stars. It is also the image driver for the imSim simulation program we are working on here at Duke.

To use it logon to cosmology-01 and then (like before if you have logged out):
source /var/phy/project/lsst/
Then type:
setup galsim
Then, copy this directory:


somewhere into your area. If you did things correctly you can go into the directory and type:
 It will run and then put a ‘fits’ file in the ‘output’ directory that you can look at with ds9.  If that worked you are ready to start.

I would like you to read and try to understand  You could play with it by writing your own program to understand it, or make a ipython notebook or something.   Once you have done this I would like you to run and understand in the same way if you can, and

You might need to look things up or ask questions to understand some of the things. Don’t be shy about asking.

Then, if you have done all of those I would like you try something which will put together all of the knowledge that you have learned so far.  It might be some work to figure it out. But, when you finish you will know how to use the program.

I would like you to try to make a simple version of this:
There is enough information in the text to figure out how to make this.  But one thing: there is no easy way to make a galaxy on its “edge” so just make galaxies from the front with about the correct size and make them the proper distance apart.