The SKAMP Project & Virtual Observatory Considerations 29 Jan 2003 ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 1
The SKAMP Project & Virtual Observatory Considerations Outline: • SKA (Square Kilometre Array) overview • SKAMP (SKA Molonglo Prototype) Project o Aims & Overview o Status • SKAMP VO Considerations o Data & data-pipeline 29 Jan 2003 ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 2
The SKA will be 100 times more sensitive than present radiotelescopes VLA and Arecibo were such large advances that collecting area unchanged for decades… Need technology shift to progress! 29 Jan 2003 ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 3
Probing the distant universe HST VLA SKA In the past few years, optical telescopes have begun to probe the ‘normal’ galaxy population to z~3 29 Jan 2003 ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 4
Proposed SKA Specifications (SKA Technical Workshop, 1997) 29 Jan 2003 ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 5
SKA Design Considerations • Large collecting area for high sensitivity (1 km 2), 100 x sensitivity of current VLA. • Array elements (stations) distributed over a wide area for high resolution (needed to avoid confusion at very faint flux levels). • For good uv plane coverage (especially for HI observations), stations can’t be too sparse. SKA will be an international US$1 billion project 29 Jan 2003 ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 6
SKA antenna concepts US ATA Australia Luneburg Lenses Canada Large reflector Dutch phased array 29 Jan 2003 China (Arecibo like) Australia cylindrical paraboloid ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 7
The Square Kilometre Array Molonglo Prototype (SKAMP) Goal: To equip the Molonglo telescope with new feeds, low -noise amplifiers, digital filterbank and FX correlator with the joint aims of: • developing and testing SKArelevant technologies and • providing a new capability for low-frequency radio astronomy in Australia 1. 6 km ··· 44 antenna segments ··· ··········································· ············ 1 2 3 ········· 44 45 ············· 88 FX Correlator ~90 Inputs ~4000 Baselines ~2000 Frequency Channels 89 90 Interference Mitigation Antennas Data Processing, Storage and Serving 29 Jan 2003 ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 8 12 m
The Square Kilometre Array Molonglo Prototype (SKAMP) Team: Anne Green 1, Tim Adams 1, John Bunton 2, Duncan Campbell-Wilson 1, Aaron Chippendale 1, 3, Bevan Jones 4, Michael Kesteven 3, Dimiter Nedialkov 1, Andrew Parfitt 2, Elaine Sadler 1, George “Ñima” Warr 1, 3 [1] School of Physics, University of Sydney [2] Telecommunications and Industrial Physics, CSIRO [3] Australia Telescope National Facility, CSIRO [4] Argus Technologies 29 Jan 2003 ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 9
SKAMP Funding • $25, 000 USyd Sesqui seed funding for SKAMP Stage I • $1. 7 million MNRF funding for components & technician for SKAMP Stage II & III • $240, 000 ARC Linkage Project for ultrawideband linefeed for SKAMP Stage III Awaiting outcome (late March ’ 03): • $192, 000 ARC Linkage (CSIRO) 3 -year Post. Doctoral Fellow for Control & Pipeline system 29 Jan 2003 ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 10
From MOST to SKA Prototype Current Survey (1997 -2003): The Sydney University Molonglo Sky Survey (SUMSS), imaging the whole southern sky (δ < -30°) at 843 MHz to m. Jy sensitivity with 45” resolution (i. e. similar to NVSS). [843 MHz, 12 hr synthesis, 2. 7° diameter field] 1. 6 km ··· 44 antenna segments ··· ··················································· FX Correlator ~90 Inputs ~4000 Baselines ~2000 Frequency Channels 29 Jan 2003 12 m Next: Use existing telescope as SKA testbed and science facility: • Frequency agility: 300 -1420 MHz (250 MHz BW) • Multi-beaming (3 independent fanbeams) • 2048 Spectral channels, 4000 baselines • Continuous uv coverage • Wide field of view (1. 5°-1420 MHz, 8°-300 MHz) ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 11
From MOST to SKA Prototype A three stage approach… I. 2003 Narrowband continuum correlator (843 MHz, 4 MHz bandwidth, 88+8 stations = 4, 560 baselines) [In parallel with SUMSS] II. 2004 Increase bandwidth and add spectral capability (800 -850 MHz, 4, 560 baselines, 2, 000 frequency channels) [Use existing linefeed] III. 2006 New linefeed + enhanced correlator (300 -1420 MHz, 250 MHz bandwidth, 4, 560 baselines, 2, 000 frequency channels) 29 Jan 2003 ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 12
Continuous uv coverage gives excellent image quality 750 m 1. 6 km (Bock et al. 1999) §Continuous uv coverage from 90 m to 1. 6 km in 12 hr synthesis §SKA will also have fully-sampled uv data 29 Jan 2003 ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 13
Stage I correlator will enable high-dynamic range imaging MOST imaging dynamic range is 100 -200: 1 (similar to intrinsic dynamic range of VLA) Use of selfcalibration on VLA has enabled imaging dynamic ranges of 105 -106: 1 A correlator will allow use of selfcalibration methods on MOST 29 Jan 2003 Current imaging dynamic range of MOST limits imaging of faint sources (eg young supernova remnants) near bright sources (eg Galactic Centre) ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr (MGPS Green et. al. ) 14
Stage II FX correlator enables spectroscopic measurements 800 -850 MHz 2, 000 spectral channel FX correlator enables, eg: • Measurements of HI absorption at z = 0. 7 to 0. 8 that capitalise on the large collecting area of MOST • Measurements of the recombination lines of C and H enabling determination of constrains on the physical conditions in the ISM (Anantharamaiah & Kantharia 1999) 29 Jan 2003 (Lane 2000) Stage II enables ΩHI measurements at z = 0. 7 to 0. 8, where other methods are not well constrained ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr (Lane and Briggs 2001) 15
Stage III Science Goals: 1. Radio galaxy spectral index FX correlator: wide-band radio spectrometry Radio spectral index measurements over the range 300 – 1400 MHz are an efficient way of selecting highredshift (z>3) radio galaxies (e. g. de Breuck et al. 2000). Radio galaxy TN 0924 -2201 at z=5. 19 (van Breugel et al. 1999) 29 Jan 2003 ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 16
Stage III Science Goals: 2. High-redshift HI in galaxies log 10 Mlim (HI) (M⊙) HIPASS (500 s) (12 h) Molonglo (10 x 12 h) Typical bright spiral HI in the nearby Circinus galaxy (Jones et al. 1999) The Molonglo telescope will reach HI mass limits typical of bright spiral galaxies at z=0. 2 (lookback time ~3 Gyr), allowing a direct measurement of evolution in the HI mass function. 29 Jan 2003 ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 17
Stage III Science Goals: 3. Other science projects § Redshifted HI absorption (z=0 to 3) FX correlator (2048 channels, each 0. 2– 25 km/s) § OH megamasers § Galactic recombination lines (H, C) Pointing agility § Rapid response to GRBs Independent fan beams Optional 64 fanbeams within main beam § Monitoring programs (SETI, pulsars etc. ) 29 Jan 2003 § Pulsar searches (high sensitivity, wide field of view) ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 18
SKA Molonglo Prototype Technologies § Multibeaming § Wide instantaneous field of view § Digital beamforming § Wide-band FX correlator (2, 048 channels, 3, 828 baselines) “Mega MOST” ~1 million frequency-image points on sky § Frequency and pointing agility § Wide-band line feeds and LNAs § Cylindrical antenna prototype § Adaptive null steering and adaptive noise cancellation (RFI) § Remote operation Collecting area 18, 000 m 2 ~= 1% of SKA (i. e. equivalent to 1 SKA station) 29 Jan 2003 ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 19
Status of SKAMP Stage I Continuum Correlator • Conceptual design complete • Input intermediate-frequency signal tap and SUMSS isolation circuit complete and in production • Digitiser, delay and fringe-stopping unit: o Internal FPGA design 60% complete o Circuit board design 20% complete (ADC prototyped) • Correlator: o Internal FPGA design 80% complete o Circuit board design 50% complete • Work commenced on control & pipeline system 29 Jan 2003 ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 20
VO Considerations – SKAMP produces masses of raw data Cylindrical Parabolic Collectors (Two collinear 778 m x 12 m) 300 -1420 MHz Feed and LNA (7, 400 feeds, 14, 800 LNAs) Delay line beamforming (Over 9 feeds) Analog to Digital Converter (1, 600 8 bit 250 MHz BW ADCs) SKAMP III SKAMP Stage I continuum correlator dumps 4, 560 baselines per second for one (4 MHz wide) frequency band. SKAMP I gives 1. 6 GB/12 hr observation. 400 GB/s raw data rate = 17 PB/12 hr obs Digital delay beamforming (80 10 m x 10 m patches) Digital filterbank (160) (Two polarisations @ 250 MHz/patch) FX Correlator (3, 160 baselines, 2, 048 channels) Independent fanbeams (3 within 9 feed field of view) 500 MB/s per beam 52 MB/s for 1 s correlator dump period = 2. 2 TB/12 hr obs 0. 8 MB/s for 60 s correlator dump period = 37 GB/12 hr obs Interference mitigation for low-earth-orbit satellites may 29 Jan 2003 21 require SKAMP & VO Considerations - G. "Ñima" Warr (imaging, spectrometer, searching. . . )ANITA Workshop -0. 1 s or faster correlator dumps. Signal processing & storage
SKAMP Data Pipeline • Intention is to create an automated data pipeline. • Stage I data to be stored in RP-FITS format utilising modified ATNF software routines. • Initially MIRIAD will be used to invert the visibility data into images for comparison with SUMSS images. • Aaron Chippendale investigating high dynamic range imaging and use of AIPS++ routines to invert data. • We will investigate implementing interference mitigation algorithms in the pipeline. Daniel Mitchell has done some preliminary calculations on the correlator dump time required for this. 29 Jan 2003 ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 22
The End 29 Jan 2003 ANITA Workshop - SKAMP & VO Considerations - G. "Ñima" Warr 23
Скачать презентацию The SKAMP Project Virtual Observatory Considerations 29
f7eb086657111b180d774aec08ed46b9.ppt