The Observing Project 2019: Photometric redshifts of galaxies
We will be using the iTelescope network of robotic telescopes for this project. You will work in pairs to obtain your data only, but will do your own analysis and write your own report. (See Queen's academic integrity page for more information). The project is divided into two parts, A and B, and instructions are available below. Ananthan Karunakaran is the TA for the course and can assist with the observing project, as can your prof, Theresa Wiegert. Ananthan should first be contacted by email at a.karunakaran at queensu.ca if needed and he will also maintain an office hour on Fridays from 9:30 to 10:30 in room 358 Stirling Hall. Once you have formed a group, choose a group name and email Ananthan to get your login to the iTelescope network. The observing period goes until the next full moon makes it impossible to make faint observations... (and after if necessary!) Next full moon is October 13, so if you can get at least the majority of your observations done before October 6, this would be optimal. A progress report after Part A is expected (see below, worth 5%), deadline October 23.
For this year's project, we will attempt to measure a photometric redshift of a galaxy of own choice. This is done by observing your target in five SDSS bands, deriving the photometry (in essence you get 5 flux data points), compare to known spectral energy distributions (SEDs) of galaxies, discern a shift of the spectrum and calculate the redshift.
In essence, this is an exercise in picking a suitable object, carrying out observations, doing analysis, including a suitable error/uncertainty analysis of your result, and finalizing everything in a scientific report. Also, in formulating questions along the way to get on with the research. The full project is worth 25% of your final grade.
Stuff to read up on: spectroscopic and photometric redshifts, optical wideband filters (SDSS and Johnson/Cousins, depending on which ones are utilized with your telescope), optical observations, calibrations (you will need to describe calibrations having been done at the observatory to your data, in your report).
Bonus: - Create a colour image to brag with, by merging the bands (can be done in for example gimp, there are tutorials online on how to accomplish this, for example this (stage 2 and onwards, things may have changed in newer versions of the software)). -Make one more observation with a grating (available on T17) to observe your own spectrum in order to calculate the spectroscopic redshift to compare with (may not be time for this, so this is for the really keen student..).
Part A: The observations
Planetarium software: If you do not already have planetarium software on your computer, the first step is to download and install this aid. Stellarium is free software which has linux, mac, and windows versions. Spend a little time exploring this software. It will help you to see the location in the sky of the planet (or any other object), when it rises and sets at a given location, how high it gets in the sky, whether it is too close to the moon to be easily observed, etc. Learn how to set your location and time, especially, as well as how to search for specific objects.
Choosing your target: This could be a time consuming exercise, but follow a few steps and it should be ok. The NASA/IPAC extragalactic catalogue (NED) will be your friend. Here you can type in galaxy names or do searches for galaxies based on their Hubble types, distances, redshift, coordinates, you-name-it. You will of course find a plethora of data, including measured redshifts, SEDs, images in different filters, to compare with. While this is ok for the final write-up, I would like you to consider your galaxy as unexplored territory, and use the data available, at this step, only to find your "goldilocks" object for your observations. There is a delicate balance here - you want a galaxy sufficiently far away to have a measurable redshift. But at the same time you need it to be nearby enough to be able to observe it in reasonable exposure times (and nearby/large enough to make a pretty picture, if so inclined, but far enough to fit within your field-of-view of the telescope...). (Note that while spiral galaxies may be more aesthetically pleasing, it is actually easier to find a good fit between photometric and spectroscopic redshifts for elliptical galaxies (see paper by e.g. Margoniner and Wittman). We will provide SEDs for both.)
The iTelescope network is an international network of amateur telescopes (located at professional sites) that are available to the public 'for hire'. The Dept. of Physics, Engineering Physics, and Astronomy has bought points on this system so that students can use the telescopes without charge. On the website, explore the telescope options by clicking on the 'telescope' tab at the top. There are telescopes in Spain, Australia, New Mexico and California. Choose between the five telescopes which have the filters that are requires (4-5 wideband optical filters, U, B, V, R, I): T17, T30 (Australia), T11, T21 (New Mexico) and T7 (Spain). Once you have chosen a telescope, it'll be easier to use the same one for all of your observations, because the scale on the CCD will always be the same and so will the necessary exposures (if you opt to make a spectroscopic measurement using a grating (see further info below), you may need another telescope, but that would be ok - the spectrum does not need to match any photometry). Ensure that you make a note of the telescope's technical specifications, especially the Long/Lat of the observing site, the field of view (FOV) and the scale, i.e. the number of arcseconds/pixel.
You will be given a username and login by the TA. Log in and go through the video tutorials so as to understand the system and learn how to set up a reservation and an observing plan.
Set up the observations: When you log in, you will log in twice: once to access the iTelescope network, and then once again to log in to the specific telescope that you have chosen. You can now see what reservations have already been made by others. Make sure that you know the rise and set times for your target from this observing site. This can be done in Stellarium by setting your 'location' to be the location of the telescope and adjusting the time as needed. Warning! Times at the telescope (iTelescope) are local times (i.e. the time as if you are physically at the telescope). Check how Stellarium understands the times (e.g. standard, daylight).
Knowing the magnitude of your target, try to estimate a suitable exposure time so that your signal to noise will be high enough (aim for 10 times the noise, higher is better. This may require some reading up and checking with TA/lecturer....
Now make a reservation a time period each upcoming night. iTelescope will only allow you to do this for the next 4 nights in a row, but you may need more nights. This means that you'll have to repeat the process of making reservations several times. The reservation system requires 15 minutes as a minimum (you will likely need more). You are competing with other people for time so its best to make your reservations without delay.
Next go to the Plan Generator and set up your observing plan (give it a name, (e.g. photz_John.txt). Choose your filter (one per observation) (no binning) for a few images of different exposure times -- and then repeat the above -- and save your plan. Then go back and edit your reservations so that they access the name of your plan. For the time, leave some time for the telescope to run through some calibrations and slew to the source at the beginning from wherever it is pointing previously (6 minutes or so, this may be telescope specific).
Checking your data as it comes in: You will receive an email as to whether the observing has successfully been carried out each night. Remember, these are REAL telescopes obtaining data in real time. It *could* be cloudy!
Download your data (blue 'Download my data' tab on the home page after the first login, OR, click on the link that you will be sent after any successful observation). After unzipping, the files you want will start with 'Calibrated'. Make sure that you know what this means!! The calibrated images have already been corrected for bias, flat fielding and thermal counts. A paragraph explaining what each of these corrections does is expected in the final write-up!
The files are in 'FITS' format. Fits stands for 'flexible image transport system' and is the standard for almost all research-based astronomical observations and many amateur images as well. The file starts with a 'header' which is just text and can be read normally. The header contains a plethora of information about the observations, including the exposure time, the UT date of the observation, what filter was used, the latitude and longitude of the site, the object name, name of the observer, information about the calibration, etc. Following the readable header is a binary file, so if you keep trying to read the file, you will see only gibberish.
To check your fits files, you will need fits-reading software (see below). Although jpgs are sent to you by email along with your notification of a successful observation, the stretch (i.e. the numerical scale) will not be ideal for you to see the galaxy as you'd like, plus you won't be able to make any useful flux measurements. Depending on how you set your stretch and field of view, foreground stars should also be visible across the field. There are a variety of fits readers available. Options for data reduction include Siril, DS9, or IRIS (to be clarified by the TA, see Part B). At this stage, you just want to see that things have worked and that you have an image of a galaxy in each band you have observed. The iTelescope network telescopes centre the source of interest very nicely, exceptions being if something has occasionally slipped at the telescope.
Submit a progress report: Due Wednesday October 23rd, 2019 The progress report should contain an image that clearly shows your galaxy, in a band of your choice. It should contain a table of information indicating what observations have been obtained, e.g. date, duration, filters, etc. Specify which telescope was used, its site, latitude and longitude, telescope aperture and CCD information such as field of view and scale. Include any other information that may be pertinent, such as the relative location of the moon and whether the observations failed because of weather or other reason. Please be specific. If you indeed have had failed observations, then the information (date, telescope, etc) should be in your table alongside the good observations, but with a flag indicating that they failed and why. The progress report should be no more than 2 pages.
The progress report is worth 5% out of the 25 total marks for the project. Pay attention to the details - this will pay off at every stage along the way!
Part B: The reductions and the analysis To be updated, please check back!
Congratulations! You have now reached the fun part! :) This part of the project requires that you make measurements from the images. TO BE UPDATED!
Your TA for the course, Ananthan gave a tutorial on how to carry out the analysis and determine the remaining part of the project on November 13th. PDF's of the presentations given at the tutorial are available here: Theresa's overview. Ananthan's more detailed presentation.
You have your data, a galaxy of your choice observed in 4 or 5 bands. The first stumbling block to tackle, is how to convert the units (data numbers, DN, i.e. counts per pixel) to flux density. In order to do that, there are a few steps to take (see tutorial presentations for more detail):
- If you haven't yet, get familiar with the Header - see more information on fits files and header above in part A. - Find a software that will help you - see list of software above. A good starting point is DS9, but apparently the version for Windows is not functioning as well as it should (for saving images). The linux and apple versions seem fine, so if you have a linux emulator of sorts, that might work better. You can still check your data and the headers with the windows version though. Iris seems to be the best bet of software to do the remainder of the work. If you are Python savvy, there are many packages available for doing the analysis as well. - Many headers of optical data will include a phrase with information on how to convert the DN to magnitudes. This is not the case for these telescopes (that would have been to easy, I guess!), so we have to figure out this conversion ourselves. The best way to do that is to compare foreground stars in your field with already known magnitudes. http://astrometry.net/ will be useful for converting your pixel locations to more useful RA and dec coordinates.
Rather than just measuring one star, measure a few stars in the field around the galaxy, to get an average. Be careful not to scale the magnitudes as if they are fluxes.
The final step is to compare your flux density values in the different bands to those of spectral energy distribution models for your type of galaxy. Coleman Wu and Weedman 1980 have a well used set. If your galaxy is a starburst, there are also starburst galaxy SEDs available. You can also check adsabs and the internet for more recent SEDs to compare with. Note, this can be done by hand, or by software, although most software are set up to do many (thousands) of galaxies at once (e.g. BPZ by Benitez). Calculate a photometric redshift. Compare it to tabulated value at NED https://ned.ipac.caltech.edu/ if available (very likely). To be added: SED files, colour files for the colour-redshift check.
Final Report is due Dec. 3, 23:59, 2019. Make sure to have your analysis done in good time before, i.e. Nov 28, to leave enough time for the report.
Please write up the final report as you would a lab, but this project is more in-depth than a laboratory report would be. The sections would therefore be: abstract, introduction, observing data and analysis, results, discussion, and conclusions. There are good examples to check up at the ads service here: https://ui.adsabs.harvard.edu/classic-form
Introduction: Do some reading about photometric redshifts and your galaxy. Introduce some known information and provide references. When is it normally useful to do photometric redshifts rather than the more accurate spectroscopic ones? Observing data and analysis: (you can make sub sections to this section - and in any section as you see fit) This should contain all of your data in tabular form (i.e. what you submitted for the progress report), as well as all the steps that you went through to reduce/measure the data. Be quantitative and be clear about the steps, even those that you didn't do yourself (e.g. what were the calibration steps that happened before you downloaded the data). Were there any issues that caused problems? Specify. Include error analysis. Indicate the name of your partner. The data analysis and write-up are your own! Do not share anything beyond the collection of the data. Results: What did you find? Double check the objectives of the project (top of this page). Discussion: Discuss the results. Does your photometric redshift agree with the known value or not? Agreement means that the error bars on your value overlap with the known value. If they do not agree (a null-result is also a result), offer a believable reason. What is your overall assessment of what you accomplished for this project? Can you suggest ways in which what you have done could be improved? Conclusion: Should be fairly obvious what to write here.