UC Riverside

Dust attenuation refers to the absorption and scattering of light by interstellar dust particles within a galaxy. This effect, which depends on wavelength, is also known as dust reddening due to its more pronounced impact on shorter wavelengths. Studying dust attenuation is important in the field of galaxy evolution as it helps astronomers to gain insight into various aspects of galaxy formation such as accurate measurements of galaxy properties, star formation history, metallicity, and chemical evolution, galaxy morphology, and classification. Dust extinction/attenuation curves are used to express the dependency of dust reddening on wavelength. I use the spectroscopic data from SDSS, MOSDEF, and MOSDEF/LRISsurveys to constrain the nebular and stellar dust attenuation curves and explore their variations with physical properties of the local and high-redshift (z ∼ 2) star-forming galaxies.

My dissertation aims to examine how nebular and stellar attenuation curves relate to the physical properties of a galaxy including stellar mass, star formation rate, and metallicity. To accomplish this, I utilize techniques such as the Balmer decrement and reconciling various Hα− and SED-based star formation rates. The motivation for examining nebular attenuation curves stems from evidence suggesting a more significant reddening for nebular emission lines compared to the stellar continuum. This disparity could be due to the presence of dust grains with distinct size and mass properties in nebular regions, resulting from strong radiation fields around massive stars. Additionally, the dust/star geometry might differ between nebular and stellar regions. We use the spectroscopic data with the availability of the first four Balmer emission lines from the SDSS survey and derive the nebular attenuation curve for a sample of 78,340 galaxies. Our results suggest that the nebular curve does not exhibit variations with respect to stellar mass, star formation rate, or metallicity.

The stellar dust attenuation curve plays a crucial role in modeling stellar populations within galaxies. These models are essential for understanding the formation and evolution of galaxies and require accurate accounting for dust attenuation effects on various stellar populations. Employing a sample of 412 star-forming galaxies with MOSFIRE optical spectra and BPASS models, we identify optimal model combinations for reconciling Hα and SED-based SFRs, finding sub-solar metallicity populations with SMC reddening provide the best agreement. We also explore stellar dust attenuation curve variations with stellar mass in 124 galaxies using Keck/LRIS far-UV spectra, revealing consistent average metallicities and the SMC curve as the best match for SFRs across both low- and high-mass galaxies. Another focus of my dissertation is to test whether the Hα-to-UV luminosity ratio (L(Hα)/L(UV)) is a reliable tracer of bursty star-formation histories (SFHs) of star-forming galaxies. Verifying the reliability of the Hα-to-UV ratio in tracing burstiness is crucial for accurately characterizing the star formation history of galaxies, interpreting observational data, and refining our understanding of galaxy formation and evolution. We analyze L(Hα)/L(UV) for 310 star-forming galaxies in two redshift bins from the MOSFIRE Deep Evolution Field (MOSDEF) survey. Using CANDELS/3D-HST imaging, we construct star-formation-rate surface density (ΣSFR) and stellar age maps and examine far-UV spectra from a 124-galaxy subsample. Our results show no significant evidence of bursty star formation based on ΣSFR distributions within a galaxy. We identify two populations with low and high average L(Hα)/L(UV) ratios but find no variations in age-sensitive FUV spectral features. Thus, we cannot conclusively confirm the reliability of the L(Hα)/L(UV) ratio in tracing burstiness for ensembles of star-forming galaxies at z ∼ 2. We introduce alternative tracers of recent star-forming activities, such as the equivalent widths of SiIV λλ1393, 1402, CIV λλ1548, 1550 P-Cygni, and HeII λ1640 stellar wind features, which are less susceptible to uncertainties known to impact the reliability of the L(Hα)/L(UV) ratio. These tracers provide valuable insights into the properties of massive stars, stellar winds, and their surrounding environments. By using multiple indicators of recent star formation, we can gain a more comprehensive understanding of the star-forming activities in galaxies and reduce potential biases introduced by using a single metric like the L(Hα)/L(UV) ratio.