I’m very happy to be posting again to the How-To-Guide. We’ve made a number of updates to Quench data and Quench Tools. Before I launch into Part 3 of the Guide, here are the recent updates:
- The classification results for the 57 control galaxies that needed replacements have been uploaded into Quench Tools.
- We’ve applied two sets of corrections to the galaxies magnitudes: the magnitudes are now corrected for both the effect of extinction by dust and the redshifting of light (specifically, the k-correction).
- We’ve uploaded the emission line characteristics for all the control galaxies.
- We’ve uploaded a few additional properties for all the galaxies (e.g., luminosity distances and star formation rates).
- We corrected a bug in the code that mistakenly skipped galaxies identified as ‘smooth with off-center bright clumps’.
In Part 3 we’re going to use the results from the classifications that you all provided to see if there’s anything different about the post-quenched galaxies that have merged or are in the process of merging with a neighbor, and those that show no merger signatures.
The figure below is of one of my favorite post-quenched galaxies with merger signatures. Gotta love those swooping tidal tails!
Let’s get started!
Step 1: Because of the updates to Tools, first clear your Internet browser’s cache, so it uploads the latest Quench Tools data.
Step 2: Copy my starter dashboard with emission line ratios ready for play.
- Open my Dashboard and click ‘Copy Dashboard’ in the upper right. This way you can make changes to it.
- In this Dashboard, I’ve uploaded the post-quenched galaxy data.
- I also opened a Table, just as you did in Part 2 of this How-To-Guide. I called the Table ‘All Quench Table’.
- In the Table, notice how I’ve applied a few filters, by using the syntax:
filter .’Halpha Flux’ > 0
- This reduces the table to only include sources that fulfill those criteria.
- Also notice that I’ve created a few new columns of data, just as you did in Part 2, by using the syntax:
field ‘o3hb’, .’Oiii Flux’/.’Hbeta Flux’
- That particular syntax means that I took the flux for the doubly ionized oxygen emission line ([0III]) and divided it by the flux in one of the Hydrogen emission lines (Hbeta).
- This ratio and the ratio of [NII]/Halpha are quite useful for identifying Active Galactic Nuclei (AGN).
- It’d be really interesting if we find that AGN play a role in shutting off the star formation in our post-quenched galaxies. A major question in galaxy evolution is whether there’s any clear interplay between merging, AGN activity, and shutting off star formation.
Step 3: Create the BPT diagram using the ratios of [OIII]/Hb and [NII]/Ha.
- BPT stands for Baldwin, Phillips, and Terlevich (1981), among the first articles to use these emission line ratios to identify AGN. Check out the GZ Green Peas project’s use of the BPT diagram.
- Click on ‘Tools’. Choose ‘Scatter plot’ in the pop-up options.
- In the new Scatterplot window, choose ‘All Quench Table’ as your ‘Data Source’.
- For the x-axis, choose ‘logn2ha’. For the y-axis, choose ‘logo3hb’.
- Adjust the min/max values so the data fits nicely within the window, as shown in the figure below.
- Remember that you can click on the comb icon in the upper-left of the plot to make the menu overlay disappear.
- Do you notice the two wings of the seagull in your plot? The left-hand wing is where star forming galaxies reside (potentially star-bursting galaxies) while the right-hand wing is where AGN reside. Our post-quenched sample of galaxies covers both wings.
Step 4: Compare the BPT diagram for post-quenched galaxies with and without signatures of having experienced a merger.
- To do this, you’ll need to first create two new tables, one that filters out merging galaxies and the other that filters out non-merging galaxies.
- Click on ‘Tools’. Choose ‘Table’ in the pop-up options.
- In the new Table window, choose ‘All Quench Table’ as the ‘Data Source’. Notice how this new table already has all the new columns that were created in the ‘All Quench Table’. That makes our life easier!
- Look through the column names and find the one that says ‘Merging’. Possible responses are ‘Neither’, ‘Merging’, ‘Tidal Debris’, or ‘Both’.
- Let’s pick out just the galaxies with no merger signatures.
- Under ‘Prompt’ type:
filter .Merging = ‘Neither’
- If you scroll to the bottom of the Table, you’ll notice that you now have only 2191 rows, rather than the original 3002.
- Call this Table ‘Non-Mergers Table’ by double clicking on the ‘Table-4’ in the upper-left of the Table and typing in the new name.
- Now follow the instructions from Step 3 to create a BPT scatter plot for your post-quenched galaxies with no merger signatures. Be sure to choose ‘Non-Mergers Table’ as the ‘Data Source’.
- You might notice that this plot looks pretty similar to the plot for the full post-quenched galaxy sample, just with fewer galaxies.
What about post-quenched galaxies that show signatures of merger activity? Do they also show a similar mix of star forming galaxies and AGN?
- To find out, create a new Table, but this time under ‘Prompt’ type:
filter .Merging != ‘Neither’
- The ‘!=’ syntax stands for ‘Not’, which means this filter picks out galaxies that had any other response under the ‘Merging’ column (i.e, tidal tails, merger, both). Notice how there are 505 sources in this Table.
- Now create a BPT scatter plot for your ‘Mergers Table’.
- Make sure this plot has a similar xmin,xmax,ymin,ymax as your other plots to ensure a fair comparison.
- You might also compare histograms of log(NII/Ha) for the different subsamples.
What do you find? Do you notice the difference? What could this be telling us about our post-quenched galaxies?!
Before you get too carried away in the excitement, it’s a good idea to compare the post-quenched galaxy sample BPT results against the control galaxy sample.
This comparison with the control sample will tell you whether this truly is an interesting and unique result for post-quenched galaxies, or something typical for galaxies in general. You might consider doing this in a new Dashboard, as I have, to keep things from getting too cluttered. In that new Dashboard, click ‘Data’, choose ‘Quench’ in the pop-up options, and choose ‘Quench Control’ as your data to upload. Now repeat Steps 1-4.
Do you notice any differences between your control galaxy and post-quenched galaxy sample results? What do you think this tells us about our post-quenched galaxies?
Stay tuned for Part 4 of this How-To-Guide. I’d love to build from your results from this stage, so definitely post the URLs for your Dashboards here or within Quench Talk and your questions and comments.
It was amazing how quickly the new Quench classifications were completed. We posted them on Friday and you were already done by Sunday morning. Wow, that’s awesome! This means we can turn our full attention to making sense of the data. And we need your help!
In Part 1 of this How-To-Guide to data analysis within Quench, you learned how to use Tools, our analysis platform, and were inspired (or so I hope) about ways to play with the data as you read the background literature about post-quenched galaxies and galaxy evolution.
In Part 2 of this How-To-Guide, we’re going to help you navigate using Tools to compare results from galaxies *you* classified with the rest of the post-quenched galaxy sample.
You’re 12 small steps away from your first comparison plot between your galaxies and the full sample… let’s get started!
Step 1: Enter Tools and log in using your Zooniverse login information.
Step 2: Choose ‘Quench’ in the pull-down menu in the upper-left, next to the words ‘zootools’. Now click ‘Create Dashboard*’ in the upper-right, and give it a name, like: ‘My Data in Context’.
Step 3: Click ‘Data’ in the upper-left and choose ‘Zooniverse’ in the pop-up options.
Step 4: In the window that pops up, choose ‘Recents’ or ‘Collections’. Your choice.
If you classified galaxies in quench.galaxyzoo.org, they’ll be accessible through ‘Recents’. Choose the max number possible. If you created a Collection of interesting galaxies in Quench Talk or want to look at someone else’s Collection, you can access them by clicking ‘Collections’.
I’ve created a Dashboard* in Tools called ‘Example: My Data in Context’. Take a look and, if you’d like, you can even make edits by copying it into your Tools environment.
In my Dashboard ‘Example: My Data in Context’, I chose ‘Collections’. I love #Quencher SUMO_2011’s Collection of ‘Blue’ galaxies from Quench. If you go to that URL, the Collection ID is listed after the final ‘/’ in the URL. In this case, the Collection ID is CGSS00000x. I inputted that ID into the pop-up window in Tools, in the box next to ‘Enter Collection Id:”. I then clicked on ‘Import Data’.
Step 5: Now that you have your galaxies’ information imported into the Dashboard, it’s time to play with them. Click on ‘Tools’ in the upper-left and choose ‘Table’ in the pop-up options.
Step 6: In your Table window, choose ‘Zooniverse-1’ in the pull-down menu under ‘Data Source’. Now the Table knows to work with that set of data.
Step 7: As in Part 1 of this How-To-Guide (https://blog.galaxyzoo.org/2013/08/23/quench-boost-a-how-to-guide-part-1/), you’ll make a new column that has color information about your galaxies. You do this by subtracting the brightness of your galaxy in one filter from the brightness of your galaxy in another filter.
In the open space under ‘Prompt’ in your Table, write: field ‘My Galaxies Color u-r’, .u-.r
If you scroll to the right in your table, you’ll see that you created a new column of information, called ‘My Galaxies Color u-r’.
Step 8: Click ‘Tools’ in the upper-left and choose ‘Scatterplot’ in the pop-up options.
Step 9: In your Scatterplot window, choose ‘Table-2’ in the pull-down menu under ‘Data Source’. Now the Scatterplot knows to work with the Table, which includes your new column with Color information.
Step 10: Choose ‘log_mass’ for the X-axis and ‘My Galaxies Color u-r’ for the Y-axis. Recent star formation is seen strongly in the u-band while older stars dominate the r-band. The color, u-r, tells you about the star formation history for each of your galaxies. Check out this post for more details.
Step 11: How do your galaxies compare with the full sample of post-quenched galaxies? To answer this, we redo the steps 3-10 above, but for the post-quenched galaxy sample.
- Click on ‘Data’ in the upper-left and choose ‘Quench’ in the pop-up options.
- Click on ‘Quench Sample’ in the pop-up window.
- Click on ‘Tools’ in the upper-left and choose ‘Table’ in the pop-up options.
- In the new Table window, choose ‘Quench-4’ in the pull-down menu under ‘Data Source’. This loads the Quench Sample into that Table.
- In the open space under ‘Prompt’ in your Table, write: field ‘Quench Galaxies Color u-r’, .u-.r
- Click on ‘Tools’ in the upper-left and choose ‘Scatterplot’ in the pop-up options.
- In the new Scatterplot window, choose ‘Table-5’ in the pull-down menu under ‘Data Source’.
- Choose ‘log_mass’ for the X-axis and ‘Quench Galaxies Color u-r’ for the Y-axis.
- Zoom in on the data, for example, choosing Xmin: 7, Xmax: 12, Ymin: 1, and Ymax: 4.
Step 12: Place your two scatterplots side by side. For a fair comparison, make sure the x- and y-axis range is the same for both plots, otherwise the stretch might skew your analysis. I tend to make the axes ranges in the plot showing My Galaxies match the plot showing the Quench Sample.
What do you notice about your subsample of post-quenched galaxies compared to the full sample? Do they occupy a particle sub-space within the plot? Or are they randomly distributed throughout the quench space?
The figure below shows what you’ll see if, like me, you uploaded SUMO_2011’s Collection of blue galaxies. You’ll notice that all of the blue-collection galaxies are way bluer (closer to the bottom of the plot, near values u-r = 1.5) than the full post-quenched galaxy sample (which spread from u-r values of 1 to u-r values of 3.5 and higher). This is a reassuring reality check given what you see visually when you look at the color of the galaxies. The plot also tells us that since these blue galaxies have such low values of ‘u-r’, they’ve had more recent star formation than most of the post-quenched galaxies.
In looking at these two plots side-by-side, I wondered: Why are there so few massive post-quenched galaxies (log_mass > 11) with bluer colors (u-r < 2.0)? If we compare our post-quenched galaxies with our control galaxies, do I see any difference? Specifically, are there massive (log_mass > 11) control galaxies with bluer colors (u-r < 2.0)? If there are, what might that be telling me about our post-quenched galaxy sample?
Stay tuned for Part 3 of our How-To-Guide for taking part in the analysis phase of the research process. If you have suggestions for what you’d like to learn more about, please post here. Thank you all, and keep on Quenching!
*Dashboard is the place within Tools (tools.zooniverse.org) for volunteers to observe, collect, and analyze data from Zooniverse citizen science projects.