LOFAR Science and Computing

1
LOFARScience and Computing

• Huub Röttgering (rottgeri_at_strw.leidenuniv.nl),
  Ger de Bruyn,
• Rob Fender, Jan Kuijpers
• Barthel, Braun, Butcher, Eric Deul Falcke,
  Garrett, Groot, van Haarlem, Ed van den Heuvel,
  Thijs van der Hulst, Israel, Michiel van der
  Klis, Koopmans, Miley, Morganti, Tom Oosterloo,
  Zaroubi, Schaye, Schilizzi, Stappers, Timmermans,
  Vermeulen, Edwin Valentijn, Verheijen, Marco de
  Vos, van der Weygaert, Wijers.

2
LOFAR

• Unique frequency range
• ? 10-240 MHz
• Unprecedented sensitivity
• Baselines 400 km
• Angular resolution 1 arcsec at 200 MHz
• Multi-beaming
• up to 8 beams

LOFAR opens up the last unexplored
wavelength region
Cabinet allocated 52 Meuro for LOFAR
3
Today

• Science
• Status project
• Software and computing

4
Four Key Science Areas in Astrophysics
5
Radio Sources are good probes of blackhole,
galaxy and cluster formation
Cygnus A a nearby radio source

• Formation and evolution of
• massive blackholes
• massive galaxies
• clusters of galaxies

The most distant sources are bright at LOFARs
low frequencies
6

• During epoch of reionization copious amounts of
  21 cm radiation
• Extremely sensitive maps from combination 100
  days of data
• Questions when, ionizing source, distribution?

7
Transient Universe

• All sky monitor
• 2 km central region
• Every 0.5 seconds
• Large Field of View
• Examples
• Gamma-ray bursts
• Exo-planets
• Galactic neutron stars/black holes

8
Cosmic ray air-showers in the earth atmosphere

• Importance
• Most important contribution to the energy density
  of gas in our Galaxy
• Origin?
• LOFAR
• Very high time resolutionat dipole level
• directions at an accuracy 1-2 degrees
• Clustering of events on the sky?
• Identification with known sources?

9
LOFAR

• The project

10
Timeline

• 1997 Feasibility studies
• 2002 User requirement review
• 2003 Preliminary Design Review
• 2004 Test Station
• 2005 Critical design review, start procurement
• 2006 Inner part array
• 2008 Full array

11
Two phases

• 1. With current allocation
• Maximum baselines 100 km
• Budgets covers
• high and low band antennas
• Connectivity (fibers etc.)
• Computers
• Operational software
• 2. With German/EU money
• Maximum baselines 400 km

12
Challenges

• Radio interference
• Spatial and temporal variations of the ionosphere
• But opens up direct studies of ionospheric
  structure
• Enormous data rates (25 Tb/s)
• Stepwise reduction

13
LOFAR Antenna
14
(No Transcript)
15
(No Transcript)
16
24h-movie of the sky at 30 MHz 200 frames,
one per 7 minT0.2s, B10 kHz, N25 dipoles
Full cross correlation matrix obtained Beam
forming for the whole sky Noise 2000
Jy Resolution 70
North

West
17
ITS image at 30 MHz N 60 dipoles T6 sec ,
B40 kHz CLEANED
Cyg A
Cas A
North Polar Spur
NORTH
Virgo A
18
LOFARSoftware telescope
19
Core Functionality

• Beamformation at station level
• Signal transportation to core
• Analysis for cosmic rays
• Interference rejection
• Correlation of signals
• (samples the Fourrier transform of image on sky)
• Calibration
• Ionosphere
• Beam patterns
• Map making
• (lots of 3-d transforms)
• Recalibration
• Deconvolution
• Analysis of final data-cubes

20
Development of (software) techniques to use the
standard LOFAR products for specific science goals

• Groningen - Reionisation (vd Hulst, Valentijn)
• Combination of 100 nights of data and search for
  the signals
• Leiden Extragalactic Surveys (Deul, HR)
• Planning, combining maps, source finding, quality
  control
• Amsterdam Transient Sources (vd Heuvel, vd
  Klis)
• Production of transient source list
• Nijmegen Cosmic Ray Sources
• Coincident detection

21
Challenges and computer sciences

• Overall system design
• Internet connectivity challenge
• Ultimately 25 Tb/s to core
• Real time processing and scheduling
• Smart data bases
• LOFAR data and products
• Links to Virtual Observatory
• Enormous computing power
• Embedded processing
• High performance computing (IBMs Blue Gene)
• Efficient deconvolution and optimization
  algorithms
• Interference rejection and calibration
• Vizualization/pattern recognition
• Interaction large database lt-gt specific
  applications
• Grid-based workflow management
• Smaller and larger science center

22
Challenge Keeping scientific progress in pace
with the enormous datastreams
M87/Virgo A 90 cm VLA Owen etal. A low
resolution and low sensitivity radio map