ASTR 421 - Science Presentation Tips

Transcript

1. Hello ASTR 421 👋 Q&A Getting ready for f inal presentations Andy Tzanidakis // atzanida@uw.edu
2. Announcements 👋 - Homework #5 due March 2 🙈 - I’ve been super busy, I’ll be posting HW #3 solutions Friday - I’ll be grading HW #4 this weekend. Looks like y’all did great 😌 I’m giving an AoT talk tonight, if you like Time Domain astronomy & telescopes 
 the talk will be live streamed on YouTube.com (search Astronomy on Tap Seattle) - I’ve been really enjoying chatting with you all about HW, keep it coming!
3. APD, Feb. 23, 2022
4. Magnitudes on Photographic Plates This is called photographic magnitudes Harvard, CFA Diameter of the exposed stars can be used 
 as a proxy to the brightness. Experienced observers also were also able to estimate the magnitudes to ~0.1 mag 
 using calibration stars with visual magnitudes 

5. Science Presentation Tips 🚀
6. Why 🤔
7. Why 🤔 Good presentation = Good Discussion = Productive Science Happy astronomers!
8. Title Essentials - Question, statement, try to be original
9. Title Essentials - Question, statement, try to be original “Origin of Diffuse Stellar Clouds substructure in the Milky Way Galaxy: In-situ or Accreted?”
10. Title Essentials - Question, statement, try to be original “Origin of Diffuse Stellar Clouds substructure in the Milky Way Galaxy: In-situ or Accreted?” “The Messy Galactic Disk of the Milky Way” “Wiggles & Wobbles in the Milky Way”
11. Title Essentials - Question, statement, try to be original “Origin of Diffuse Stellar Clouds substructure in the Milky Way Galaxy: In-situ or Accreted?” “The Messy Galactic Disk of the Milky Way” “Wiggles & Wobbles in the Milky Way” - Don’t make it too long or complicated A Comparative Study of Artificial Neural Networks Using Reinforcement Learning and Multidimensional Bayesian Classification Using Parzen Density Estimation for Identification of GC-EIMS Spectra of Partially Methylated Alditol Acetates on the World Wide Web
12. Title Essentials - Name, course, Date - Since you’re talking about a research paper. Make sure to 
 add the arXiv number & title of the paper somewhere
14. Introduction - Motivate, motivate, motivate, motivate, motivate - In a concise way, clearly state the f indings of the paper! I like to personally use the TLDR of the paper Open questions, context, mystery, why
17. Main Body - It’s very hard to summarize a paper in 5 mins & that’s okay. Instead focus on the main message & the methods that you want your audience to learn - Quick run through on how they set up the experiment/paper/analysis - Break down dif f icult concepts into visual aids to help you and the 
 audience understand the methods of the paper - You don’t need to mention every single detail/caveat of the paper
18. 4 Andy Tzanidakis - Journal Club Presentation 2021 15 Want to model the PRF of each pixel fmi,j
19. 4 Andy Tzanidakis - Journal Club Presentation 2021 16 SAP photometry Modeling the PRF
20. 4 Andy Tzanidakis - Journal Club Presentation 2021 16 s = ′  l ∑ i=0 fi,t SAP photometry fm = SXw Assume that s is the true flux (i.e without the effects of PRF) model flux diagonal matrix 
 values of s design matrix Vector weights Modeling the PRF
21. 4 Andy Tzanidakis - Journal Club Presentation 2021 17 Modeling the PRF X = CBV0,0 CBV1,0 r0 c0 . . . 1 CBV0,1 CBV1,1 r1 c1 . . . 1 . . . . . . . . . . . . . . . 1 CBV0,N CBV1,N rN cN . . . 1 Design matrix is capable of capturing the known systematics 
 that can change the shape of the PSF Cotrending Basis Vectors 
 captures focus change systematics 
 & other prominent systematics 
 across the channel (Kinemutchi et al. 2012) Terms from a 
 2D polynomial 
 captures the effects of drifting Allows each pixel 
 to have a different mean Flux-position correlations {sr, sc, scr, s}
22. 4 Andy Tzanidakis - Journal Club Presentation 2021 18 ̂ wi = K−1 wi ((SX)TK−1 fi fi) K−1 wi = (SX)TK−1 fi (SX) Fit each TPF to fi nd weights 
 and build the PRF fl ux model Modeling the PRF Note: blank pixels in PRF model 
 have been removed due to saturation Hedges et al. 2021 Result of a PRF model, where the PRF 
 shape essentially captures the systematic 
 effects added to the design matrix *covariance is just the photometric uncertainties
23. Main Body - It’s very hard to summarize a paper in 5 mins & that’s okay. Instead focus on the main message & the methods that you want your audience to learn - Quick run through on how they set up the experiment/paper/analysis - Break down dif f icult concepts into visual aids to help you and the 
 audience understand the methods of the paper - You don’t need to mention every single detail/caveat of the paper - Identify the key f igure of the paper and discuss its interpretation
24. 4 Andy Tzanidakis - Journal Club Presentation 2021 25 Validate Color Dependence Catalog of EB’s (Matson et al. 2017) fEB500nm fEB700nm Flux model of each EB at each the peak 
 of each wavelength. Assuming blackbody emission 
 for each star. TEBi TEBi+1 Standard deviation in the residual eclipse depths Hedges et al. 2021 3 correlation between theoretical 
 prediction in color variability vs PRF shape 
 change! σ
25. Conclusions - What have we learned? What has the community learned? Why is this important? Are there next steps (usually in the Conclusions) - If you’re talking about a technique or theory, what are the applications? If you’re talking about observations, what comes next?
26. Practical Tips
27. 😟
30. VS Gaia Collaboration, 2018 Gaia Collaboration, 2018
31. VS Flux model of each EB for bins of wavelength. Assuming blackbody emission 
 for each star. Produce a f lux model of each EB for bins of wavelength. Assuming a blackbody emission for each star
32. Vivas et al. 2017 Cite Work
33. Ballasteros et al. 2017 Instead of Words Use Figures(*) & Schematics
34. Journal Club Presentation, 2022 3 TL;DR The disk-halo interface is full of defuse & thin stellar substructures that are consistent with a disk origin. This note discovers new substructures at low galactic latitude using kinematics. (not to scale, lol) A possible Scenario
35. Questions & Answers - Be polite when asking a question. If possible, acknowledge the speaker and ask your question respectfully - It’s okay if you don’t know the answer. It’s not productive to make things up, instead say something like: “that’s a really good question, and unfortunately I don’t know the answer, but I’m very interested in looking this up after this talk”