The Far-Ultraviolet Extragalactic Legacy (FUEL) Survey: Hubble Far-UV Images and Catalogs of the Extragalactic Legacy Fields

Abstract

We present far-ultraviolet (FUV) images and catalogs from the Hubble Space Telescope (HST) Advanced Camera for Surveys/Solar Blind Channel (ACS/SBC) F150LP (about 1600 Angstrom) of three extragalactic fields: GOODS-S, GOODS-N, and COSMOS. The data comprise 365 orbits of high-resolution imaging of 151 pointings covering an area of 44.7 square arcmin to typical depths of FUV about 28.7 AB (3-sigma, 0.5 arcsec diameter aperture). We provide a new model of the spatially varying dark "glow" created from all 365 orbits of data, and scale and subtract it from all pointings. We provide drizzled image mosaics, weight maps, and exposure time maps matched in coordinates and pixel scale to the Hubble Legacy Fields (HLF) frame, and the original COSMOS tiles. Galaxy photometry is measured within isophotes defined with existing deep Hubble F606W or F814W optical filters. We detect 1068 galaxies and provide catalogs of all optical detections, including matched IDs to existing 3D-HST and CANDELS catalogs. The redshift distribution of FUV-detected galaxies peaks at z about 0.6 and declines to z = 1.2, where the Lyman limit shifts redward of any filter transmission. These data fill the redshift gap of high-resolution far-UV imaging between z about 0 and z > 1, enabling studies of star-forming regions, dust properties, the FUV extragalactic background, and Lyman continuum emission from galaxies at z > 1.2.

Figures (10)

• Figure 1: Footprints of SBC/F150LP observations in the COSMOS, GOODS-N, and GOODS-S fields, overlaid on CANDELS 2011ApJS..197...35G2011ApJS..197...36K ACS/WFC F814W exposure maps (gray scale). Blue squares mark ACS/SBC far-UV pointings, red outlines indicate the footprint of the ACS optical data from the GOODS Survey 2004ApJ...600L..93G, and black outlines show wide-field optical HST/ACS imaging in COSMOS and GEMS 2004ApJS..152..163R. The scale bar in each panel indicates 30 arcminutes.
• Figure 2: Histogram showing the number of ACS/SBC pointings (left axis) and cumulative area (right axis) as a function of exposure time in orbits for COSMOS (red), GOODS-N (green), and GOODS-S (blue). Most pointings are 2-3 orbits in depth, with 16 pointings reaching depths of 5-6 orbits.
• Figure 3: A small region in GOODS-South imaged for two orbits with the ACS/WFC F606W filter (left) and the ACS/SBC F150LP filter (right). Each panel spans $\sim13"\times13"$ (about $\sim 1/4$ of a single image), and the scale bar indicates $5"$. Two sources circled in blue are detected at high significance in the far-UV. The two circled sources have far-UV S/N of $\sim$19 (brighter) and $\sim$4 (fainter). Although the far-UV image contains fewer sources overall, it is very deep ($\sim28.7$ AB), depending on exposure time and proximity to dark glow; $3\sigma$ within a $0.5"$ diameter aperture). Centroid positions were determined for each high-S/N, compact source. A comparison of the pixel positions of source centers in the two images is used to measure the pre-alignment offsets, which are typically $\sim1"$ in translation and $<1^\circ$ in rotation.
• Figure 4: Left: Stacked ACS/SBC exposures (uniform dark current subtracted and sources masked), revealing the residual background from detector glow plus diffuse sky; bad and hot pixels are excluded. Middle: Ninth-order 2D polynomial fit to the left panel, providing a smooth master dark glow model. In both the stacked (left) and the model (middle), the maximum of the dark glow is $\approx15\times$ the level of diffuse sky background around the edges of the frame. Right: Pixel-wise percent-difference map between the model (middle) and the data (left). For display, a NaN-aware $20\times20$ boxcar smoothing is applied. This demonstrates that the model is tracking the shape of the dark glow to less than 7% across peak of the dark glow signal.
• Figure 5: Matched cutouts in F150LP (left) and F606W (right). From top to bottom: two spirals, an elliptical, and an irregular galaxy.