Blue Compact Dwarf Galaxies
Results from:
Janowiecki & Salzer 2014, ApJ, 793, 109 (arXiv) optical/NIR surface photometry for structure parameters
Janowiecki et al. 2017, ApJ, 836, 128 (arXiv) SED-fitting for stellar populations & star formation history
When they were first discovered, Blue Compact Dwarf galaxies (BCDs) appeared to be isolated HII regions without any host galaxy (Sargent & Searle 1970), but deeper observations have consistently revealed underlying older stellar populations in all BCDs (Loose & Thuan 1986, Aloisi et al. 2007). Definitions of BCDs in the literature have varied from the original “extragalactic HI regions” of Sargent & Searle (1970), to dwarf galaxies with “several knots and some low surface brightness fuzz” in Sandage & Binggeli (1984), and as generic as “low-mass galaxies experiencing a starburst” (Lelli et al. 2012). This vague and inconsistent definition has resulted in a wide variety of
objects being included in samples of BCDs.
Nomenclature debates notwithstanding, we have studied a sample of BCDs which span the full range of "definitely BCD" (where everyone agrees) to "maybe not a BCD" (where the definition is fuzzy). Our goal is not to define BCDs, but to study extreme galaxies which are compact and have lots of star formation to see what they can tell us about galaxy evolution and deviations from normal scaling relations.
So far, this work has included two main components, comparing BCDs and non-BCDs:
--stellar structures of the underlying galaxy host (2014 paper)
--stellar populations and star formation histories (2017 paper)
The main figures/results from each paper are shown below:
In Janowiecki & Salzer (2014), we characterize the underlying "host" galaxy of BCDs and other star forming dwarf galaxies by thoroughly masking the star-forming parts of the galaxies and only fitting isophotes to the smooth outskirts of the galaxy. Thus, we are fitting the regions of old stars which are less affected by the current intense star formation.
Here we show BCDs as solid dots and triangles, and our comparison sample as open squares. The axes show NIR (H filter) structural parameters:
M_H is the absolute H magnitude (aka directly related to the stellar mass)
alpha_H is the exponential scale length (in kpc) of the isophotal fit to the outskirts of the galaxies
mu_0,H is the central surface brightness of that exponential fit
Note that the BCDs are more compact and reach higher central surface brightnesses than most of the comparison sample.... and this is in terms of the OLD stars in the OUTSKIRTS! That's neat because it shows that some of the BCDs would still be unique (structurally) even if their current star formation faded. This argues that some BCDs are not just star-bursting dwarf galaxies, but that there may be a compact class of dwarf galaxies out there which perhaps can have especially strong bursts of star formation.
In Janowiecki et al. (2017), we used panchromatic photometry (from UV to IR) we fit the Spectral Energy Distributions (SEDs) of a sample of BCDs and "normal" galaxies from the Local Volume Legacy (LVL) survey. While these SED fits generate many output parameters, the plot below shows the simple relation SFR and stellar mass.
BCDs from our sample are shown as blue dots while irregular/spiral galaxies from LVL are shown as open grey circles. (Brown triangles are LVL dEs; black stars are LVL galaxies sometimes considered BCDs). Note that the left panel shows SFR from H-alpha observations and the right panel shows the 50-Myr averaged SFR from our best-fitting SED SFHs. As has been shown before, most galaxies lie on a so-called "main sequence", but BCDs are offset above that. The H-alpha SFRs show even stronger offsets than the 50-Myr averaged SFRs, suggesting that the star formation enhancement in BCDs may be especially recent.
Work is ongoing to identify populations of quiescent (post-burst) BCDs through their underlying compactness and from stellar population analysis which may show evidence of past BCD-like bursts.