z ~ 1.5 STARBURST-DRIVEN IONIZED GAS OUTFLOWS
Cosmic Dawn Center, Copenhagen — Advisor: Dr. Steven Gillman
I was a DAWN-IRES Scholar this past summer at the Cosmic Dawn Center through the University of Copenhagen with Dr. Steven Gillman using integral field spectroscopy observations of high-redshift star-forming galaxies. I analyzed the starburst-driven gas outflows of these z ~ 1.5 galaxies from KMOS. Learn more about the program here and my personal experience here. If you are an undergraduate considering applying, feel free to reach out to me with any questions!
Although the program was remote (due to COVID-19), I will be visiting the Cosmic Dawn Center next summer and will have the opportunity to present my research at the 2022 DAWN Summit. I am also presenting my research at the American Astronomical Society meeting this January in Salt Lake City, Utah. Here is the abstract I submitted:
We present evidence for starburst-driven ionized-gas outflows in z ~ 1.5 star-forming galaxies using observations from the K-Band Multi Object Spectrograph (KMOS) Galaxy Evolution Survey (KGES). The sample of 187 galaxies, selected based on S/N and absence of AGN features, has a median stellar mass and extinction-corrected H star formation rate of M* =(1.0 + 0.2) 10^10 M and SFR = (20 + 2) M/yr, respectively. To detect and quantify the broad emission-line features of the warm ionized interstellar medium in the sample, we remove the large-scale dynamics of each galaxy thus enabling us to stack the galaxy integrated rest-frame optical spectra. To constrain the relationship between a galaxies main-sequence properties and the presence of ionized gas outflows, we used a running median of mass and star-formation rate to define the bin width for stacked spectra and used SFR distance from the main sequence at z = 1.5 as a different approach to binning the sample. We develop a Gaussian profile fitting routine in combination with LMFIT to search for broad, underlying emission features within the stacked spectra. The FWHM and other properties of the broad- and narrow-components of the H, [SII], and [NII] features were then quantified and we calculated the significance of the two-component fit versus a one-component fit without the presence of gas outflows. We found that the broad, underlying components increase in FWHM with both increasing stellar mass and increasing star-formation rate. These galactic scale outflows regulate galaxy formation and better constraining their physical properties are important in linking our cosmic noon sample with observations of the local universe.
Here are a few relevant figures from the research:
​
​
BACKGROUND
The galactic scale outflows from stellar winds and supernovae explosions regulate galaxy formation and both observed in the local universe and expected to be important in understanding the epoch of peak star formation (or cosmic noon) around z ~ 1-3. It is therefore necessary to better constrain the physics properties of these star-formation-driven outflows at high redshift to better understand stellar feedback.
Cosmic high noon at z ~ 2 (P. Madau & M. Dickinson)
THE SAMPLE
Distribution of the KMOS sample in the M-SFR plane. The solid gray line indicates the main sequence of star forming galaxies at z ~ 1.5 from Whitaker et al. 2014 and the dashed lines represent SFR offsets.
DATA REDUCTION
The narrow-band Hα image of CDFS-32626 (left) and the velocity map of CDFS-32626 with the kinematic axis represented by the black dashed line (right). This procedure was repeated for all 288 galaxies in the sample. The size of the KMOS observation PSF for CDFS-32626 is shown as the filled grey circle.
The observed position-velocity diagram and corresponding spectrum of CDFS-32626 in the Hα region (summed and interpolated across kinematic axis) (upper-left and -right). This is repeated for the velocity-subtracted data (lower-left and -right).
COMPOSITE SPECTRA
There seemed to be some broadening of the composite stack emission lines that likely due to the ionized gas outflows of these galaxies. To statistically test whether a two-component (narrow- and broad- component) fit was better than a one-component fit, I needed to develop a fitting routine.
The composite stack of all 187 velocity-subtracted galaxy spectra with possible broadening of emission lines (Hα, [NII], and [SII] doublets) from starburst-driven gas outflows.
DEVELOP GAUSSIAN FITTING ROUTINE
FITTING THE COMPOSITE STACK
The rest-frame composite of all 187 KMOS observations with the corresponding one-component and two-component Gaussian fits.
M - SFR BINNED STACKS
The 187 galaxies divided into four bins based on M and SFR running medians. The corresponding narrow- and broad-component fits are shown for each bin. Since the broad component characterizes the ionized gas outflow velocity, we can conclude that the velocity tends to increase with increasing M and SFR.
If you have any questions about this research, feel free to contact me via email!