Intro to Lea Hagen (2020)

A summary of my research (past, present, and future), and a bit about my non-science self. Presented at ISM*@ST group meeting on 13 Apr 2020.

Transcript

1. Who is THAT? An intro to Lea Hagen 13 Apr 2020
2. My travels ✯ Alpine ✯ Carnegie Obs. ✯ Harvey Mudd College ✯ Lowell Obs. ✯ Penn State ✯ STScI
3. Projects • Current/planned work ✴ Dust with the BEAST ✴ UV attenuation curves in the Local Volume with Swift/ UVOT ✴ UVOT observations of newly discovered dwarfs ✴ Calibrating UVOT for extended sources • Past science ✴ UGC1382: a giant low surface brightness galaxy ✴ SFR density evolution in CDF-S • Non-science things
4. The BEAST
5. Wavelength (μm) Log Flux Star Star+Dust Bayesian Extinction And Stellar Tool: Model ALL the stars & dust!
6. METAL (Roman-Duval+19): BEAST runs Table 1: BEAST Model Parameters Parameter Unit Description Min Max Resolution Prior log(t) years stellar age 6.0 10.13 0.1 flat SFR log(M) M stellar mass 0.8 2.0 variable Kroupa IMF log(Z/Z ) · · · stellar metallicity 2.1 0.3 0.3 flat AV mag dust column 0.01 10.0 0.05 flat RV · · · dust average grain size 2.0 6.0 0.5 peaked at ⇠3 fA · · · dust mixture coe cient 0.0 1.0 0.2 peaked at 1 d [LMC] kpc distance 40 60 2.5 flat d [SMC] kpc distance 47 77 2.5 flat 3.2 Source density maps When running the BEAST on stars in M31 from a survey similar to METAL, Gordon et al. (2016) found that the noise properties of stars vary considerably with the local source density (SD), especially for the lower-resolution NIR bands. Therefore, in order to properly incorporate the noise, we must independently consider sources within di↵erent local SDs. We create a SD map using sources with F475W magnitudes between 26 and 15 (the range over which the catalogs are complete) with a pixel size of 500. We consider together the pixels with values between 0 and 1 sources/arcsec2, 1 and 2 sources/arcsec2, and so on; we will refer to these throughout the section as SD bins.2 We split the catalogs into NSD sub-catalogs based on the sources’ SD bins. If a sub-catalog is especially long (> 6000 sources), we sort it by flux, and then split it into smaller 6000-source files. As described in Section 3.5, splitting the catalog in this manner has the e↵ect of improving the computational e ciency of the modeling. Resource usage: For each METAL field, constructing the SD map and splitting the catalog only takes ⇠5 minutes and 200 MB of memory. Together, the map file and sub-catalogs use 100 MB of disk space. 3.3 Artificial star tests Modeling 33 fields with XSEDE (29,000 TB-hours) Science: maps of dust parameters, dust emissivity, variations of dust grain properties F275W F475W F814W
7. Spectroscopic UV extinction curve candidates from BEAST modeling GALEX FUV IC1613 HST proposals ANGST+LUVIT HST imaging (PIs Dalcanton, Gilbert) Imaging of star-forming region Field A/B: proposed imaging Field C: existing imaging
8. Swift/UVOT
9. Swift Satellite UVOT XRT BAT BAT - Burst Alert Telescope ➝ detect GRBs XRT - X-Ray Telescope UVOT - UV/Optical Telescope ➝ GRB follow-up Gehrels+04
10. UVOT & GALEX: a comparison GALEX UVOT GALEX: UVOT: • Higher sensitivity • probes far-UV • 1.2° diam. FOV • 5" PSF • More NUV filters • 17'×17' FOV • 2.5" PSF
11. Swift UV Survey of M31 25' 5.5 kpc Exposure time: 31 hours Total Tiles: 24
12. SUMaC: Swift UV Survey of the Magellanic Clouds 0.5° 520 pc Exposure time: 1.8 days 5σ uvm2 depth: 21.5 mag Point sources: 250,000 Total Tiles: 50 D = 60 kpc
13. SUMaC: Swift UV Survey of the Magellanic Clouds 0.5° 440 pc Exposure time: 5.4 days 5σ uvm2 depth: 21.5 mag Point sources: 1,000,000 Total Tiles: 150 D = 50 kpc
14. SOLV: Swift Observations of the Local Volume • 465 galaxies from 11HUGS (Kennicutt+08): representative sample across mass, morphology, and star formation history • Ancillary data: GALEX, HST, SDSS, PanSTARRS, Hα, Spitzer, WISE, Herschel, ... • Swift observations: 15ks per galaxy ‣ 5ks in each of uvw2, uvm2, uvw1 ‣ typical 3σ SB: ~26 AB mag/arcsec2 ‣ total time (new+archival): 12 Ms • Science goals: ‣ Add to the legacy of existing wide-field UV imaging ‣ Constrain variation of UV attenuation curve
15. Slope (R V) 2175Å Bump How to correct light for dust attenuation? (this is not an exhaustive list!)
16. UVOT constrains the slope and 2175Å bump
17. Galaxy Modeling FUV to NIR imaging PEGASE.2 model spectra Dust: AV, RV, 2175Å bump MCMC Modeling
18. RV Sample attenuation curve map: M31 2175Å Bump
19. UVOT observations of newly-discovered local dwarfs ervations and analysis. This information is vital to deducing the inform our understanding of galaxy growth and development. m - y, n - f r r rophysical Journal Letters, 798:L21 (6pp), 2015 January 1 Tollerud et al. . Upper panels: gr color composite images of dwarfs Pisces A (left) and Pisces B (right) from pODI on WIYN. The images are 1′ tall, north is up, and east he slit for the Palomar optical spectroscopy is shown as the gray lines. Lower panels: GALEX AIS images at the same location and orientation as the upper or Pisces A (left), only NUV imaging is available, while for Pisces B (right), the image is an NUV/FUV color composite. nd r-band filters, with integration times of 600–1200 s per er target. Standard imaging reductions were performed ODI Portal, Pipeline, and Archive facility. These include btraction, flat-fielding, and alignment of individual Or- these two candidates are shown in the upper panels of Figure 1. They are also visible in the SDSS, although the SDSS catalog incompletely deblends them into a mix of stars and galaxies. Also shown in the lower panels of Figure 1 are images from the The Astrophysical Journal Letters, 798:L21 (6pp), 2015 January 1 Tollerud et al. Figure 1. Upper panels: gr color composite images of dwarfs Pisces A (left) and Pisces B (right) from pODI on WIYN. The images are 1′ tall, north is up, and east is left. The slit for the Palomar optical spectroscopy is shown as the gray lines. Lower panels: GALEX AIS images at the same location and orientation as the upper panels. For Pisces A (left), only NUV imaging is available, while for Pisces B (right), the image is an NUV/FUV color composite. the g- and r-band filters, with integration times of 600–1200 s per filter per target. Standard imaging reductions were performed by the ODI Portal, Pipeline, and Archive facility. These include bias subtraction, flat-fielding, and alignment of individual Or- thogonal Transfer Array (OTA) cells into chips. The SWarp program (Bertin et al. 2002) was used to combine the individual exposures, and DAOPHOT (Stetson 1987) was used to perform PSF-fitting photometry on stars in the field. Most of the H i clouds did not have optical counterparts with morphologies like nearby galaxies within the ∼4′ GALFA-H i beam. Those in the Sloan Digital Sky Survey (SDSS; Ahn et al. 2014) footprint show neither diffuse features like the galaxies described below, nor point source overdensities to the limit of the DR10 catalog. Similarly, our deeper pODI imaging showed neither overdensities nor red giant branch (RGB) features in the color–magnitude diagrams (CMD) down to r . 24 (an RGB tip distance > 3 Mpc) for any of the targets we observed other than the two described below. Only two objects showed nearby dwarf galaxy-like optical counterparts within the GALFA-H i beam. The pODI images of these two candidates are shown in the upper panels of Figure 1. They are also visible in the SDSS, although the SDSS catalog incompletely deblends them into a mix of stars and galaxies. Also shown in the lower panels of Figure 1 are images from the GALEX All-sky Imaging Survey (AIS; Morrissey et al. 2007). The morphology of the objects in these images and presence of detectable UV flux is consistent with both being dwarf (irregular) galaxies. Additionally, the presence of such point sources resolved in ground-based imaging implies that the galaxies are relatively nearby (.10 Mpc). In particular, Pisces A (left panel of Figure 1) shows point sources resolved enough to generate a CMD. We discuss this further in Section 3 in the context of providing a distance estimate. While the centroid of the optical (and GALEX) objects are offset by 30′′–40′′ from the H i emission, this is well within the 4′ uncertainty from the GALFA-H i beam. All other optical counterparts within the beam are less likely to be associated with the H i; they either appear stellar or are consistent with being distant background galaxies (and hence at too high a redshift to match the H i). Furthermore, the Hα emission discussed in the Swift UVOT GALEX January 1 Tollerud et al. physical Journal Letters, 798:L21 (6pp), 2015 January 1 Tollerud et al. pper panels: gr color composite images of dwarfs Pisces A (left) and Pisces B (right) from pODI on WIYN. The images are 1′ tall, north is up, and east slit for the Palomar optical spectroscopy is shown as the gray lines. Lower panels: GALEX AIS images at the same location and orientation as the upper Pisces A (left), only NUV imaging is available, while for Pisces B (right), the image is an NUV/FUV color composite. Swift UVOT GALEX Motivation: acquire deep UV imaging of dwarfs with little/no GALEX data Pilot program in 2016: 25ks for Pisces A & B (edge of local group) New data in 2018: Galaxy D (kpc) MV rh (pc) Obs (ks) Reticulum 2 32 -2.7 55 25 Eridanus 3 87 -2.0 11 25 NGC6503-d1 5270 -10.5 400 15 GALFA-dw3 7000 -12 350 20 GALFA-dw4 8000 -12 400 20
20. UVOT Calibration: scattered light and background offsets Uncalibrated scattered light map (Breeveld+10) UV images of Leo A showing donut and halo (~15% effect!)
21. UVOT Calibration: scattered light and background offsets uvw1 mosaic of M31 Standard pipeline Manually adjusting 110 individual snapshots for SL and background offsets calibration, including a full processing pipeline for extended sources, will be writte available on github with complete documentation. This will make it accessible to current and prospective Swift users more fully exploit the UVOT data archive. 5. Report on Previous Swift and Related Programs calibration, including a full processing pipeline for extended sources, will be writte available on github with complete documentation. This will make it accessible to current and prospective Swift users more fully exploit the UVOT data archive. 5. Report on Previous Swift and Related Programs
22. Past science
23. UGC1382: giant LSB galaxy -2 -1 0 1 2 arc-minute -2 -1 0 1 2 -2 -1 0 1 2 -2 -1 0 1 2 -2 -1 0 1 2 -2 -1 0 1 2 -60 -40 -20 0 20 40 60 kpc -60 -40 -20 0 20 40 60 kpc -60 -40 -20 0 20 40 60 kpc -2 -1 0 1 2 GALEX + SDSS Stripe 82 -4 -2 0 2 arc-minute -2 0 2 4 arc-minute 23 kpc -4 -2 0 2 arc-minute -4 -2 0 2 arc-minute -200 -100 0 100 200 relative velocity [km/s] flux velocity • physical properties: Distance: 80 Mpc UV/Optical radius: 80 kpc HI radius: 110 kpc Stellar mass: 8×1010 M⦿ HI mass: 1.7×1010 M⦿ Dark matter fraction: 0.95 • Green Valley (long term) • inefficient SF • low density environment • spiral arms are older HI with VLA Formed by accreting LSB dwarfs Hagen+16
24. UVOT at high redshift: Chandra Deep Field South Hoversten+09, Basu-Zych+11, Hagen+15 1' 550ks in w2, m2, w1, u Figure 8. SFRD and luminosity density for each redshift bin, compared to literature values. Data comefrom the MLE fits with fixed αand are shown both with and without a dust correction. Uncertainties include the contribution from cosmic variance. The large SFRD at z = 0.7 is due to the galaxy The Astrophysical Journal, 808:178 (9pp), 2015 August 1
25. Other things
26. Diversity and inclusion www.tiny.cc/InclusiveAstro2 • Inclusion2@stsci.edu All students, astronomers, social scientists, policy makers, and advocates: Come take part in a community discussion to build upon the 2015 Nashville Recommendations, reflect on the state of the profession, address issues affecting underrepresented groups, and envision how to improve astronomy into the 2020s. October 14–15, 2O19 Baltimore • MD Space Telescope Science Institute Ground Rules for Respectful Discussions • • ‒ • • ‒ • • ‒ • • INVISION READING GROUP • • ‒ • • ‒ • • ‒ • • Sources: the STScI Diversity in Astronomy group, Inclusive Astronomy 2015, the Banneker Institute, & Lydia X. Z. Brown 1. Lean into discomfort Talking about racism, sexism, heterosexism, cissexism, and ableism is not easy – but learning only happens when you leave your comfort zone. Try to translate discomfort into constructive questions. 2. Share the air Be mindful of how much you and those around you are speaking. If you find yourself dominating the conversation, please step back; if you have not spoken much, feel encouraged to step up. 3. Be aware of privilege and power How does your identity and status affect how you speak and listen to others? If you belong to an overrepresented group, please be even more careful not to dominate the discussion. 4. Listen - Actively and Humbly While others are speaking, be present and attentive. Avoid mentally imposing your own biases, thoughts, or opinions onto what someone else is sharing. If you paraphrase what someone else said, verify with them afterwards that you have correctly interpreted their words: “Did I get that right?” 5. Use “I” statements Speak from your own experiences and avoid generalizing; respect that others’ experiences and expertise will differ from your own. If you are discussing a friend or co-worker’s experience, avoid sharing their name or any identifying information. 6. The Vegas Rule Maintain confidentiality. What happens here, stays here; what is learned here, leaves here. 7. Intent ≠ Impact Recognize that unintended harm is still harm. 8. Ouch,” “oops” An important part of social justice conversations is making – and learning from – mistakes. If you feel uncomfortable or hurt by something that is said, you can say “ouch.” In response, the speaker should simply say “oops” and step back to allow the other person to explain what bothered them, if they feel so inclined. 9. The “Both/And” Rule Think “both/and” rather than “either/or.” Acknowledge that binaries are incomplete; leave room for complexity and avoid oversimplification. 10. Tolerate and teach; do not shame and blame Be patient with, rather than critical of, individuals’ unfamiliarity with issues. 11. Access the space as you need Please make yourself comfortable in our discussion space. You are welcome to sit in chairs, sit or lie on the floor, stand, rock, flap, spin, move around, and step in and out of the room. 12. Use the Parking Lot and the Safe Space If a discussion gets off track, feel free to “park it” and return to it later. Relatedly, if you feel uncomfortable with participation at any moment, you are welcome to remove yourself. The Invision Reading Group is a group founded to create an open and supportive space in which we can read about and discuss issues of diversity and inclusion within the STScI community. Invision Reading Group
27. Gardening
28. Assorted critters sourdough starter small human many many worms