Скачать презентацию Update on LSST GSMT Jeremy Mould Users Скачать презентацию Update on LSST GSMT Jeremy Mould Users

fca9dc594868f078244245b97c7c8dac.ppt

  • Количество слайдов: 31

Update on LSST & GSMT Jeremy Mould Users Committee October 13, 2004 1 Update on LSST & GSMT Jeremy Mould Users Committee October 13, 2004 1

GSMT SWG The GSMT SWG is a community-based group convened to: • Formulate a GSMT SWG The GSMT SWG is a community-based group convened to: • Formulate a powerful science case for federal investment in GSMT – Identify key science drivers – Develop clear and compelling arguments for GSMT in the era of JWST/ALMA – Discuss realization of key science as a function of design parameters: aperture, FOV, PSF…… • Generate unified, coherent community support 2

GSMT SWG Members Chair: Rolf-Peter Kudritzki, UH If. A SWG Members: – – – GSMT SWG Members Chair: Rolf-Peter Kudritzki, UH If. A SWG Members: – – – – Jill Bechtold -- UA Mike Bolte -- UCSC Ray Carlberg -- U of T Matthew Colless -- ANU Irena Cruz-Gonzales -- UNAM Alan Dressler -- OCIW Betsy Gillespie -- UA – – – – Terry Herter -- Cornell Jonathan Lunine -- UA LPL Claire Max -- UCSC Chris Mc. Kee -- UCB Francois Rigaut -- Gemini Chuck Steidel -- CIT Steve Strom -- NOAO 3

TMT is a fusion of 3 concepts GSMT CELT VLOT The GSMT, CELT and TMT is a fusion of 3 concepts GSMT CELT VLOT The GSMT, CELT and VLOT point design telescope concepts. 4

TMT Project FY 2004 • • • Project Office established Project manager appointed Engineering TMT Project FY 2004 • • • Project Office established Project manager appointed Engineering efforts from 4 partners integrated to provide a 'reference design' – based on the heritage of the VLOT, CELT and GSMT efforts • • Moore funds in place for D and D Phase (gift to UC & Caltech) CFI funds authorized NSF proposal submitted Key milestone: Baseline Design which will answer the following key design issues/trades – – Is the elevation axis in front of or behind the primary? Is the telescope optical configuration RC or AG? What is the focal ratio of the primary ( f/1 – f/1. 5)? What final focal ratios should be provided ( f/15 – f/22)? 5

Highest Priority Capabilities for First Light • diffraction-limited (10 mas @ 1. 6 m) Highest Priority Capabilities for First Light • diffraction-limited (10 mas @ 1. 6 m) imaging & spectroscopy – – 0. 8 - to 2. 5 -micron wavelength 1 -2 arcminutes multi-conjugate adaptive optics (MCAO) field Strehl ratio at K-band of 0. 7, constant across the field to 10%; highly-multiplexed (~1, 000 slits) • seeing-limited 100 < R < 7, 000 spectroscopy – 0. 32 - to 1 -micron wavelength range – wide (10 -20 arc-minute) field • high-spectral-resolution (20, 000 < R < 100, 000) spectroscopy – 1 - to 5 -micron – 7 - to 28 -micron 6

TMT phased implementation • optical spectroscopy with 20, 000 < R < 100, 000 TMT phased implementation • optical spectroscopy with 20, 000 < R < 100, 000 – 0. 3 microns to 1 micron • very high-contrast imaging near diffraction limit 1 to 2. 5 m – contrast ratio > 108 at q > 4 /D from bright stars • R ~3, 000 -5, 000 spectroscopy – – – • fields ≥ 5 arcminutes 0. 7 - to 2. 5 -micron sampling 0. 15 arcseconds image quality 80% enclosed energy in 0. 3 arc-sec. unit (IFU) heads or microslits ground-layer adaptive optics system (GLAO); mid-IR diffraction-limited imaging (Strehl > 0. 5, 7 m < < 28 m) over a field >30 arcseconds; 7

TMT AO modes AO mode Enables Science Mid-IR NGS AO Diffraction limited resolution > TMT AO modes AO mode Enables Science Mid-IR NGS AO Diffraction limited resolution > M Planet-forming Environments MCAO Diffraction-limited resolution in J, H, K bands over 0. 5 -1’ fields Galaxy Assembly; deconstructing stellar populations MOAO ~0. 1” resolution over 35’ fields for multi-object spectroscopy Young galaxy mass, metallicity, & star formation Ex. AO High dynamic range imaging Planet detection & characterization GLAO 0. 2 -3” resolution over 5 - Galaxy evolution 10’ fields 8

NIO • • • provided 'point design' for GSMT -- key element of TMT NIO • • • provided 'point design' for GSMT -- key element of TMT planning supports site testing (northern chile; Baja, CA; Hawaii); serves both the. GMT and TMT communities interfaces with ESO to advance technologies of mutual interest has contributed key technical and management leadership within TMT post TMT project office, NIO will – carry out two key TMT work packages (mid-IR Echelle; M 2 assembly) – continue site testing – continue ESO collaborations (level TBD following allocation of TMT workpackages) 9

GSMT vs JWST Simulated monochromatic images of the ‘Antennae’ (local starburst galaxy: 105 seconds GSMT vs JWST Simulated monochromatic images of the ‘Antennae’ (local starburst galaxy: 105 seconds integration time) Courtesy: Elizabeth Barton, GSMT SWG GSMT JWST 10

GMT alternate Magellan partners + Texas 7 x 8. 4 meter mirrors Giant Magellan GMT alternate Magellan partners + Texas 7 x 8. 4 meter mirrors Giant Magellan Telescope (GMT) 11

OIR Planning • Long Range Planning Committee (Chair: C. Pilachowski) is currently working on OIR Planning • Long Range Planning Committee (Chair: C. Pilachowski) is currently working on a roadmap for large scale facilities http: //www. noao. edu/dir/lrplan/lrp-committee. html • Where will the decision points be for public funding ? • Look forward from 2005 as far as 2030. – Two decadal surveys will occur before 2025, and these will outrank this roadmap. • The plan will show present investments – – realize the new initiatives, illustrate convergence paths, lay the basis for facility closures and transfers, and address community structural change. 12

Overview and Status Opportunities for Scientific Participation 9 October 2004 13 Overview and Status Opportunities for Scientific Participation 9 October 2004 13

LSST Partners 14 LSST Partners 14

Project Technical Status Systems Engineering – Requirements and Scope ~3 Gigapixel Camera 8. 4 Project Technical Status Systems Engineering – Requirements and Scope ~3 Gigapixel Camera 8. 4 m 3 -mirror Telescope Data Products & Management 15

Observing Simulator • Initially Created By Abi Saha • New Simulation Tool in Development Observing Simulator • Initially Created By Abi Saha • New Simulation Tool in Development – K. Cook et. al. – Foundation and Testbed for Scheduler A. Saha, NOAO 16

The LSST Key Science Drivers • Nature of Dark Energy – Image Quality – The LSST Key Science Drivers • Nature of Dark Energy – Image Quality – FWHM <0. 8 arcsec – Shape Systematics – PSF (e 1, e 2) < 0. 0001 • Solar System Map – Observing Cadence – Absolute Astrometry – Link Vectors From Multiple Epochs • Optical Transients – Observing Cadence – Data Processing – Real-Time Alerts (~30 sec delay) • Galactic Structure – Photometric Precision – 1% Internal, 2% Absolute – Astrometry 17

18 18

Mapping the Solar System: Probing the Fossil Record 19 Mapping the Solar System: Probing the Fossil Record 19

Mapping the Galactic Halo 20 Mapping the Galactic Halo 20

Weak Lensing and Cosmology • Cluster tomography – Shear used to obtain mass maps Weak Lensing and Cosmology • Cluster tomography – Shear used to obtain mass maps – Number density of clusters as function of redshift depends on density fluctuations and distance scale – Both depend on dark energy • Strauss report – Power spectrum, bispectrum, and shear cosmography 21

Special cadence to go deeper? 22 Special cadence to go deeper? 22

Evolving Optical Design T 3=3. 25, CC 2=-0. 5501 23 Evolving Optical Design T 3=3. 25, CC 2=-0. 5501 23

24 24

Wide – Deep - Fast • • • ~10 deg. 2 per Field ~7 Wide – Deep - Fast • • • ~10 deg. 2 per Field ~7 m Effective Collecting Area m~24 th per 10 sec Exposure Wide Coverage > 15, 000 Square Degrees Multiple Filters (e. g. bgriz´ - TBD) ~100+ Epochs in Each of >1500 Fields in Each Filter Over Ten Years • Accumulated Depth of 26 th Magnitude in Each Filter 25

Schedule & Milestones First Light 2004 2005 2006 2007 2008 2009 2010 2011 2012 Schedule & Milestones First Light 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Co. DR PDR CDR Design Order glass Construction Start final camera fab Optics on site Integration First Light Commissioning Operation 26

Camera • Focal Plane Array – 10 mm pixels 0. 2 arcsecond/pixel (~1/3 seeing-limited Camera • Focal Plane Array – 10 mm pixels 0. 2 arcsecond/pixel (~1/3 seeing-limited PSF) – 64 cm diameter 10 square degree FOV 3 Gpixels – Integrated front-end electronics – 16 bits/pixel, 2 sec readout time 3 GB/sec • Parallel readout Housing / Filters / Optics / Mechanisms 27

Primary Mirror Contract • Private Donor Committed to Buy LSST Mirror – University of Primary Mirror Contract • Private Donor Committed to Buy LSST Mirror – University of Arizona Borosilicate Cast Mirror – Similar to LBT Primary with Very Large Hole • Contract Approved – – Materials and Engineering Casting Optical Figuring Cell Integration and Testing 28

Telescope Structure • Initial Warren Davidson Study Complete – Long Tube – Stiff Structure, Telescope Structure • Initial Warren Davidson Study Complete – Long Tube – Stiff Structure, f(n 1)=10 hz – Relatively Light , 200 T • Preparing for Second Study – Short Tube – Open Structure – Industrial Source 29

Site Selection • First Down-Selection Completed in May 2004 • • Cerro Pachon Las Site Selection • First Down-Selection Completed in May 2004 • • Cerro Pachon Las Campanas San Pedro Martir La Palma • Study to Evaluate Satellite Data Issues • Correlating Local Data to Global Weather Patterns • Final Site (2) Selection Meeting 14 January 2005 30

Summary • • • LSST Corporation is Established The Mission is Solidifying Management Organized Summary • • • LSST Corporation is Established The Mission is Solidifying Management Organized & Vision is Clear Project Teams Developing Technical Advancement Accelerating 31