We present new multi-band imaging data in the optical (BVRI and Halpha) and near infrared bands (JHK) of 15 candidate ring galaxies from the sample of Appleton & Marston (1997). We use these data to obtain color composite images, global magnitudes an
d colors of both the ring galaxy and its companion(s), and radial profiles of intensity and colors. We find that only nine of the observed galaxies have multi-band morphologies expected for the classical collisional scenario of ring formation, indicating the high degree of contamination of the ring galaxy sample by galaxies without a clear ring morphology. The radial intensity profiles, obtained by masking the off-centered nucleus, peak at the position of the ring, with the profiles in the continuum bands broader than that in the Halpha line. The images as well as the radial intensity and color profiles clearly demonstrate the existence of the pre-collisional stellar disk outside the star-forming ring, which is in general bluer than the disk internal to the ring. The stellar disk seems to have retained its size, with the disk outside the ring having a shorter exponential scale length as compared to the values expected in normal spiral galaxies of comparable masses. The rings in our sample of galaxies are found to be located preferentially at around half-way through the stellar disk. The most likely reason for this preference is bias against detecting rings when they are close to the center (they would be confused with the resonant rings), and at the edge of the disk the gas surface density may be below the critical density required for star formation. Most of the observed characteristics point to relatively recent collisions (<80 Myr ago) according to the N-body simulations of Gerber et al. (1996).
We present the results obtained from an objective search for stellar clusters, both in the currently active nuclear starburst region, and in the post-starburst disk of M82. Images obtained with the HST/ACS in F435W(B), F555W(V), and F814W(I) filters
were used in the search for the clusters. We detected 653 clusters of which 393 are located outside the central 450 pc in the post-starburst disk of M82. The luminosity function of the detected clusters show an apparent turnover at B=22 mag (M_B=-5.8), which we interpret from Monte Carlo simulations as due to incompleteness in the detection of faint clusters, rather than an intrinsic log-normal distribution. We derived a photometric mass of every detected cluster from models of simple stellar populations assuming a mean age of either an 8 (nuclear clusters) or 100 (disk clusters) million years old. The mass functions of the disk (older) and the nuclear (younger) clusters follow power-laws, the former being marginally flatter (alpha=1.5+/-0.1) than the latter (alpha=1.8+/-0.1). The distribution of sizes (Full Width at Half Maximum) of clusters brighter than the apparent turn-over magnitude (mass>2E+4 Mo) can be described by a log-normal function. This function peaks at 10 pc for clusters more massive than 1E+5 Mo, whereas for lower masses, the peak is marginally shifted to larger values for the younger, and smaller values for the older clusters. The observed trend towards flattening of the mass function with age, together with an over-abundance of older compact clusters, imply that cluster disruption in M82 is both dependent on the mass and size of the clusters.
We present the results of applying new object classification techniques to difference images in the context of the Nearby Supernova Factory supernova search. Most current supernova searches subtract reference images from new images, identify objects
in these difference images, and apply simple threshold cuts on parameters such as statistical significance, shape, and motion to reject objects such as cosmic rays, asteroids, and subtraction artifacts. Although most static objects subtract cleanly, even a very low false positive detection rate can lead to hundreds of non-supernova candidates which must be vetted by human inspection before triggering additional followup. In comparison to simple threshold cuts, more sophisticated methods such as Boosted Decision Trees, Random Forests, and Support Vector Machines provide dramatically better object discrimination. At the Nearby Supernova Factory, we reduced the number of non-supernova candidates by a factor of 10 while increasing our supernova identification efficiency. Methods such as these will be crucial for maintaining a reasonable false positive rate in the automated transient alert pipelines of upcoming projects such as PanSTARRS and LSST.
