Search

Scientific analyses commonly compose multiple single-process programs into a dataflow. An end-to-end dataflow of single-process programs is known as a many-task application. Typically, HPC tools are used to parallelize these analyses. In this work, we investigate an alternate approach that uses Apache Spark—a modern platform for data intensive computing—to parallelize many-task applications. We implement Kira, a flexible and distributed astronomy image processing toolkit, and its Source Extractor (Kira SE) application. Using Kira SE as a case study, we examine the programming flexibility, dataflow richness, scheduling capacity and performance of Apache Spark running on the Amazon EC2 cloud. By exploiting data locality, Kira SE achieves a 4.1× speedup over an equivalent C program when analyzing a 1TB dataset using 512 cores on the Amazon EC2 cloud. Furthermore, Kira SE on the Amazon EC2 cloud achieves a 1.8× speedup over the C program on the NERSC Edison supercomputer. A 128-core Amazon EC2 cloud deployment of Kira SE using Spark Streaming can achieve a second-scale latency with a sustained throughput of $\sim$ ∼ 800 MB/s. Our experience with Kira demonstrates that data intensive computing platforms like Apache Spark are a performant alternative for many-task scientific applications.

21 pags., 12 figs., 4 tabs., The See Change survey was designed to make z > 1 cosmological measurements by efficiently discovering high-redshift Type Ia supernovae (SNe Ia) and improving cluster mass measurements through weak lensing. This survey observed twelve galaxy clusters with the Hubble Space Telescope (HST) spanning the redshift range z = 1.13-1.75, discovering 57 likely transients and 27 likely SNe Ia at z ~ 0.8-2.3. As in similar previous surveys, this proved to be a highly efficient use of HST for supernova observations; the See Change survey additionally tested the feasibility of maintaining, or further increasing, the efficiency at yet higher redshifts, where we have less detailed information on the expected cluster masses and star formation rates. We find that the resulting number of SNe Ia per orbit is a factor of ~8 higher than for a field search, and 45% of our orbits contained an active SN Ia within 22 rest-frame days of peak, with one of the clusters by itself yielding 6 of the SNe Ia. We present the survey design, pipeline, and supernova discoveries. Novel features include fully blinded supernova searches, the first random forest candidate classifier for undersampled IR data (with a 50% detection threshold within 0.05 mag of human searchers), real-time forward-modeling photometry of candidates, and semi-automated photometric classifications and follow-up forecasts. We also describe the spectroscopic follow-up, instrumental in measuring host galaxy redshifts. The cosmology analysis of our sample will be presented in a companion paper. * Based on observations with the NASA/ESA Hubble Space Telescope, obtained at the Space Telescope Science Institute, which is operated by AURA, Inc., under NASA contract NAS 5-26555, under programs 13677, 14327., Based in part on observations obtained at the Gemini Observatory, which is operated by the Association of Universities for Research in Astronomy, Inc., under a cooperative agreement with the NSF on behalf of the Gemini partnership: the National Science Foundation (United States), National Research Council (Canada), CONICYT (Chile), Ministerio de Ciencia, Tecnología e Innovación Productiva (Argentina), Ministério da Ciência, Tecnologia e Inovação (Brazil), and Korea Astronomy and Space Science Institute (Republic of Korea). Based on observations collected at the Table 4. European Organisation for Astronomical Research in the Southern Hemisphere under ESO program(s) 294.A-5025(A), 095.A-0830(A, B, C), 096.A-0926(B, C), 097.A-0442(A, B, C), and 0100.A-0851(A). G.W. acknowledges support from the National Science Foundation through grant AST-1517863, by HST program numbers GO-13677/14327.01 and GO-15294, and by grant No. 80NSSC17K0019 issued through the NASA Astrophysics Data Analysis Program (ADAP). G.A. and R.G. acknowledge support from HST program NASA HST-GO14163.002-A, and by grant No. NNH16AC25I issued through NASA ADAP. M. J. J. acknowledges support for the current research from the National Research Foundation (NRF) of Korea under the programs 2017R1A2B2004644 and 2020R1A4A2002885. H. Hildebrandt is supported by a Heisenberg grant of the Deutsche Forschungsgemeinschaft (Hi 1495/5-1) as well as an ERC Consolidator Grant (No. 770935). Support for program numbers GO-13677/14327.01 and GO-15294 was provided by NASA through a grant from the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Incorporated, under NASA contract NAS5-26555. This work was also partially supported by the Office of Science, Office of High Energy Physics, of the U.S. Department of Energy, under contract no. DE-AC02-05CH11231. Facilities: Hubble Space Telescope, Keck:I (LRIS), Keck:I (MOSFIRE), VLT, Gemini, GTC, Subaru. Software: astropy (Astropy Collaboration 2013), DrizzlePac (Gonzaga et al. 2012), iPython (Pérez & Granger 2007), Matplotlib (Hunter 2007), Numpy (van der Walt et al. 2011), SEP (Barbary 2016), scikit-learn (Pedregosa et al. 2011), SciPy

Correction of Type Ia Supernova brightnesses for extinction by dust has proven to be a vexing problem. Here we study the dust foreground to the highly reddened SN 2012cu, which is projected onto a dust lane in the galaxy NGC 4772. The analysis is based on multi-epoch, spectrophotometric observations spanning 3,300 - 9,200 {\AA}, obtained by the Nearby Supernova Factory. Phase-matched comparison of the spectroscopically twinned SN 2012cu and SN 2011fe across 10 epochs results in the best-fit color excess of (E(B-V), RMS) = (1.00, 0.03) and total-to-selective extinction ratio of (RV , RMS) = (2.95, 0.08) toward SN 2012cu within its host galaxy. We further identify several diffuse interstellar bands, and compare the 5780 {\AA} band with the dust-to-band ratio for the Milky Way. Overall, we find the foreground dust-extinction properties for SN 2012cu to be consistent with those of the Milky Way. Furthermore we find no evidence for significant time variation in any of these extinction tracers. We also compare the dust extinction curve models of Cardelli et al. (1989), O'Donnell (1994), and Fitzpatrick (1999), and find the predictions of Fitzpatrick (1999) fit SN 2012cu the best. Finally, the distance to NGC4772, the host of SN 2012cu, at a redshift of z = 0.0035, often assigned to the Virgo Southern Extension, is determined to be 16.6$\pm$1.1 Mpc. We compare this result with distance measurements in the literature., Comment: 48 pages, 13 figures. Accepted for publication in The Astrophysical Journal. The spectral time series data presented in this article can be found at http://snfactory.lbl.gov/snf/data/

We report the discovery of a multiply lensed Ly alpha emitter at z = 3.90 behind the massive cluster WARPS J1415.1+3612 at z = 1.026. Images taken by the Hubble Space Telescope using the Advanced C ...

Initial release.

We present DES13S2cmm, the first spectroscopically-confirmed superluminous supernova (SLSN) from the Dark Energy Survey (DES). We briefly discuss the data and search algorithm used to find this event in the first year of DES operations, and outline the spectroscopic data obtained from the European Southern Observatory (ESO) Very Large Telescope to confirm its redshift (z = 0.663 +/- 0.001 based on the host-galaxy emission lines) and likely spectral type (type I). Using this redshift, we find M_U_peak = -21.05 +0.10 -0.09 for the peak, rest-frame U-band absolute magnitude, and find DES13S2cmm to be located in a faint, low metallicity (sub-solar), low stellar-mass host galaxy (log(M/M_sun) = 9.3 +/- 0.3); consistent with what is seen for other SLSNe-I. We compare the bolometric light curve of DES13S2cmm to fourteen similarly well-observed SLSNe-I in the literature and find it possesses one of the slowest declining tails (beyond +30 days rest frame past peak), and is the faintest at peak. Moreover, we find the bolometric light curves of all SLSNe-I studied herein possess a dispersion of only 0.2-0.3 magnitudes between +25 and +30 days after peak (rest frame) depending on redshift range studied; this could be important for 'standardising' such supernovae, as is done with the more common type Ia. We fit the bolometric light curve of DES13S2cmm with two competing models for SLSNe-I - the radioactive decay of 56Ni, and a magnetar - and find that while the magnetar is formally a better fit, neither model provides a compelling match to the data. Although we are unable to conclusively differentiate between these two physical models for this particular SLSN-I, further DES observations of more SLSNe-I should break this degeneracy, especially if the light curves of SLSNe-I can be observed beyond 100 days in the rest frame of the supernova., Accepted by MNRAS (2015 January 23), 13 pages, 6 figures, 2 tables

We have directly observed broadband thermal noise in silica/tantala coatings in a high-sensitivity Fabry-Perot interferometer. Our result agrees well with the prediction based on indirect, ring-down measurements of coating mechanical loss, validating that method as a tool for the development of advanced interferometric gravitational-wave detectors., Final version synchronized with publication in Phys. Lett. A

Using three magnified Type Ia supernovae (SNe Ia) detected behind CLASH clusters, we perform a first pilot study to see whether standardizable candles can be used to calibrate cluster mass maps created from strong lensing observations. Such calibrations will be crucial when next generation HST cluster surveys (e.g. FRONTIER) provide magnification maps that will, in turn, form the basis for the exploration of the high redshift Universe. We classify SNe using combined photometric and spectroscopic observations, finding two of the three to be clearly of type SN Ia and the third probable. The SNe exhibit significant amplification, up to a factor of 1.7 at $\sim5\sigma$ significance (SN-L2). We conducted this as a blind study to avoid fine tuning of parameters, finding a mean amplification difference between SNe and the cluster lensing models of $0.09 \pm 0.09^{stat} \pm 0.05^{sys}$ mag. This impressive agreement suggests no tension between cluster mass models and high redshift standardized SNe Ia. However, the measured statistical dispersion of $\sigma_{\mu}=0.21$ mag appeared large compared to the dispersion expected based on statistical uncertainties ($0.14$). Further work with the supernova and cluster lensing models, post unblinding, reduced the measured dispersion to $\sigma_{\mu}=0.12$. An explicit choice should thus be made as to whether SNe are used unblinded to improve the model, or blinded to test the model. As the lensed SN samples grow larger, this technique will allow improved constraints on assumptions regarding e.g. the structure of the dark matter halo., Comment: Minor updates to match MNRAS published version. 15 pages, 7 figures. For additional info, see http://www.supernova.lbl.gov

We present a measurement of the volumetric Type Ia supernova (SN Ia) rate out to z ~ 1.6 from the Hubble Space Telescope Cluster Supernova Survey. In observations spanning 189 orbits with the Advanced Camera for Surveys we discovered 29 SNe, of which approximately 20 are SNe Ia. Twelve of these SNe Ia are located in the foregrounds and backgrounds of the clusters targeted in the survey. Using these new data, we derive the volumetric SN Ia rate in four broad redshift bins, finding results consistent with previous measurements at z > 1 and strengthening the case for a SN Ia rate that is equal to or greater than ~0.6 x 10^-4/yr/Mpc^3 at z ~ 1 and flattening out at higher redshift. We provide SN candidates and efficiency calculations in a form that makes it easy to rebin and combine these results with other measurements for increased statistics. Finally, we compare the assumptions about host-galaxy dust extinction used in different high-redshift rate measurements, finding that different assumptions may induce significant systematic differences between measurements., 11 pages, 7 figures. Submitted to the Astrophysical Journal. Revised version following referee comments. See the HST Cluster SN Survey website at http://supernova.lbl.gov/2009ClusterSurvey for control time simulations in a machine-readable table and a complete listing of transient candidates from the survey

We present a new survey strategy to discover and study high-redshift Type Ia supernovae (SNe Ia) using the Hubble Space Telescope (HST). By targeting massive galaxy clusters at 0.9 0.95, nine of which were in galaxy clusters. This strategy provides an SN sample that can be used to decouple the effects of host-galaxy extinction and intrinsic color in high-redshift SNe, thereby reducing one of the largest systematic uncertainties in SN cosmology.

We present a new measurement of the volumetric rate of Type Ia supernova up to a redshift of 1.7, using the Hubble Space Telescope (HST) GOODS data combined with an additional HST dataset covering the North GOODS field collected in 2004. We employ a novel technique that does not require spectroscopic data for identifying Type Ia supernovae (although spectroscopic measurements of redshifts are used for over half the sample); instead we employ a Bayesian approach using only photometric data to calculate the probability that an object is a Type Ia supernova. This Bayesian technique can easily be modified to incorporate improved priors on supernova properties, and it is well-suited for future high-statistics supernovae searches in which spectroscopic follow up of all candidates will be impractical. Here, the method is validated on both ground- and space-based supernova data having some spectroscopic follow up. We combine our volumetric rate measurements with low redshift supernova data, and fit to a number of possible models for the evolution of the Type Ia supernova rate as a function of redshift. The data do not distinguish between a flat rate at redshift > 0.5 and a previously proposed model, in which the Type Ia rate peaks at redshift >1 due to a significant delay from star-formation to the supernova explosion. Except for the highest redshifts, where the signal to noise ratio is generally too low to apply this technique, this approach yields smaller or comparable uncertainties than previous work., Accepted for publication in ApJ

We present Keck diffraction limited H-band photometry of a z~1.3 Type Ia supernova (SN) candidate, first identified in a Hubble Space Telescope (HST) search for SNe in massive high redshift galaxy clusters. The adaptive optics (AO) data were obtained with the Laser Guide Star facility during four observing runs from September to November 2005. In the analysis of data from the observing run nearest to maximum SN brightness, the SN was found to have a magnitude H=23.9 +/- 0.14 (Vega). We present the H-band (approximately rest-frame R) light curve and provide a detailed analysis of the AO photometric uncertainties. By constraining the aperture correction with a nearby (4" separation) star we achieve 0.14 magnitude photometric precision, despite the spatially varying AO PSF., Comment: 11 pages, 8 figures, Accepted for Publication in AJ Updated the citations, fixed typos

We present weak gravitational lensing analysis of 22 high-redshift (z >~1) clusters based on Hubble Space Telescope images. Most clusters in our sample provide significant lensing signals and are well detected in their reconstructed two-dimensional mass maps. Combining the current results and our previous weak-lensing studies of five other high-z clusters, we compare gravitational lensing masses of these clusters with other observables. We revisit the question whether the presence of the most massive clusters in our sample is in tension with the current LambdaCDM structure formation paradigm. We find that the lensing masses are tightly correlated with the gas temperatures and establish, for the first time, the lensing mass-temperature relation at z >~ 1. For the power law slope of the M-TX relation (M propto T^{\alpha}), we obtain \alpha=1.54 +/- 0.23. This is consistent with the theoretical self-similar prediction \alpha=3/2 and with the results previously reported in the literature for much lower redshift samples. However, our normalization is lower than the previous results by 20-30%, indicating that the normalization in the M-TX relation might evolve. After correcting for Eddington bias and updating the discovery area with a more conservative choice, we find that the existence of the most massive clusters in our sample still provides a tension with the current Lambda CDM model. The combined probability of finding the four most massive clusters in this sample after marginalization over current cosmological parameters is less than 1%.