Jerry Nelson

1
Science Goals and AO for the TMT

• Jerry Nelson
• University of California at Santa Cruz
• Adaptive Optics for Extremely Large Telescopes
• Paris, 2009June23

2
Outline

• Project Introduction
• Telescope overview
• Science-based metrics
• TMT key features
• Major science goals
• Science Instruments

3
Project Introduction

• Time line
• 2004 project start, design development
• 2009 preconstruction phase
• 2011 start construction
• 2018 complete, first light, start AO science
• Partnership
• UC
• Caltech
• Canada
• Japan
• NSF?
• Others?

4
Telescope Overview
LGSF launch telescope
M2 support tripod
M2 structural hexapod
Tensional members
LGSF beam transfer
M2 hexagonal ring
M2 support columns
Elevation journal
Nasmyth platform
Azimuth cradle
Laser room
M1 cell
Azimuth truss
5
TMT Optical DesignRitchey Chrétien

• M1 Parameters
• ø30m, f/1, Hyperboloid
• k -1.000953
• Paraxial RoC 60.0m
• Sag 1.8m
• Asphericity 29.3mm (entire M1)
• M2 Parameters
• ø3.1m, f/1, Convex hyperboloid,
• k -1.31823
• Paraxial RoC -6.228m
• Sag 650mm
• Aspheric departure 850 mm
• M3 Parameters
• Flat
• Elliptical, 2.5 X 3.5m

6
Segment Size
7
Primary Mirror Control System (M1CS)

• The M1CS, with the Alignment and Phasing System,
  turn the 492 individual segments into the
  equivalent of a monolithic 30 meter diameter
  mirror.
• TMT control strategy is an evolutionary
  improvement on the successful strategy used at
  the two Keck Telescopes.

SSA prototype with dummy segment
7
8
M1CS Overview

• M1CS maintains the overall shape of the primary
  mirror
• Attenuates gravity, temperature, wind, and
  vibration disturbances
• The primary mirror is aligned and phased using
  the Alignment and Phasing System (APS) every 4
  weeks or after a segment exchange.
• Look up tables are used in between calibration
  runs
• M1CS controls the global shape of the M1 using
  segment-mounted edge sensors and actuators
• Real time On-instrument Wavefront Sensors
  (OIWFS) measurements or AO system offloads will
  augment the static look up tables built using APS
  data

M1 surface error from wind disturbance
M1CS off 223 nm RMS
M1CS on 14 nm RMS
9
Science-based Metrics

• We use the time needed to make an observation as
  our metric
• Generally assume that we are observing point
  sources
• Generally assume the sources are background
  limited (most photons come from background,
  rather than source)
• In detail this is based on Kings paper that
  shows
• For these assumptions we get
• This is true for seeing-limited or
  diffraction-limited observations

10
Science Merit Function

• For seeing-limited observations
• PSS D2/image diameter2
• For diffraction-limited observations
• image solid angle varies as 1/area, so we get the
  well known PSSD4 rule
• For finite Strehl, the signal strength is reduced
  by S, but the background is not reduced, so one
  gets PSSS2 where

11
Reflectivity, emissivity, throughput

• Clearly we want the highest possible reflectivity
  of our optics
• Obvious, since PSS throughput
• In the visible r 0.9, so for 3 mirrors, net
  throughput 0.73
• But, as important, thermal emission from the warm
  optics can increase the IR background,
  particularly in K band
• When the IR sky is dark (between OH lines) the
  telescope emission can be the dominant background
  source
• Background ( warm mirrors)(1-reflectivity)
• So it can be VERY important to minimize the
  number of warm mirrors between the target and the
  IR instrument

12
(No Transcript)
13
Thermal backgrounds

• Previous graph shows the blackbody flux
• Cooling the optic by 30 reduces the flux in
  this wavelength region by a factor of 15
• Below 2µm the flux is lower than natural
  backgrounds
• These fluxes are multiplied by the mirror
  emissivities and the number of mirrors
• Observatory created backgrounds
• Three ambient temperature telescope mirrors (M1,
  M2, M3)
• NFIRAOS science path
• 1 ambient window
• 5 cold mirrors
• 1 cold beam splitter
• 1 cold window

14
K band thermal background

• In the near IR, only K band will see significant
  thermal flux from telescope
• Telescope
• 3 mirrors at 1.5 emissivity each
• Segment gaps 0.5
• Net background 0.05ambient blackbody flux
• NFIRAOS
• Net throughput 85, so emissivity 0.15
• Cooling 30 reduces flux by a factor of 15, so
• Net added background 0.01ambient blackbody
  flux

15
Field of View

• For many science programs larger field of view is
  useful
• Multiple targets
• Complex targets (galaxies, etc)
• Astrometry where reference objects are needed
• Seeing-limited unvignetted 15 arcmin FoV
• Atmospheric angular anisoplanatism limits the
  correctable field of view for AO
• One must measure the atmosphere over a sufficient
  volume to know what the angle dependent
  correction needs to be
• With lasers, one must do tomography to get this
  information
• One must have multiple deformable mirrors to make
  the added correction

16
Impact of multiple deformable mirrors

• More DMs allow greater 3-d fidelity of
  atmospheric correction, improving correction over
  larger field of view

17
TMT design path

• 30m diameter telescope
• High reflectivity optics
• Only 3 reflections to science instruments or
  NFIRAOS
• NFIRAOS cooled by 30 to reduce thermal emission
• NFIRAOS is initial AO system and can feed 3 inst
• For AO, two DMs for increased field of view
• For AO, large sky coverage enabled by using 3
  partially corrected natural stars (focus, tip,
  tilt) with 6 LGS
• 15 arcmin unvignetted field of view for
  seeing-limited
• All instruments always available in lt 10 minutes

18
Nasmyth Configuration First Decade Instrument
Suite
/IRMS
TMT GCAR, April 2009
18
19
NFIRAOS MCAO has better performance than current
systems

• Dual conjugate AO system
• Order 61x61 DM and TTS at h0 km
• Order 75x75 DM at h12 km
• Better Strehl than current AO systems
• (e.g., Keck 280-300nm WFE)

• Completely integrated system
• Fast (lt5 min) switch between targets
• gt50 sky coverage at galactic poles (w/lt2mas TT
  error)

19
20
TMT Key Science

• Nature and composition of the Universe
• Formation of the first stars and galaxies
• Evolution of galaxies and the intergalactic
  medium
• Relationship between black holes and their
  galaxies
• Formation of stars and planets
• Nature of extra-solar planets
• Presence of life elsewhere in the Universe

20
21
Science Drivers for large O/IR Telescopes3
Basic Types

• Science that you know you want to do now, but
  have discovered to be out of reach through
  experience on 8-10m telescopes.
• These tend to be what is written in design
  reference mission or science case documents
• Solving problems we do not even know about yet
• Thinking about capability space, or discovery
  space, rather than specific science cases
• Some intuition is necessary- where will the
  surprises be, what will we need to follow them
  up?
• Supporting roles and complementarity with other
  facilities on ground, in space.
• Harder to make such roles sound
  exciting/compelling BUT next-generation O/IR
  telescopes will play key role in supporting ALMA,
  JWST, CCAT, LSST, IXO (CON-X), ZEUS, etc.
• While many other facilities may not publicly
  admit that they need large O/IR telescopes on
  the ground (for the same reason), in fact,
  history suggests that they will.

21
22
TMT Detailed Science Case

• 100 page summary of TMT science case (David
  Silva, editor), completed and posted publicly in
  October 2007. (http//tmt.org)
• Developed with AURA/NOAO as full partner (US
  community interests accounted for).
• Includes science cases developed by instrument
  feasibility study teams
• From fundamental physics and cosmology, to galaxy
  and structure formation, to extra-solar planets,
  to solar system studies.

22
23
Key TMT features for Science

• 30m, f/1 primary, RC telescope, 20 field
• 30-m is a judgment about the proper balance
  between science benefit, cost, technological
  readiness, and schedule
• Filled aperture, 492 1.44m segments
• produces a more concentrated point spread
  function (PSF), improving signal-to-noise ratios
  and easing data analysis
• Integrated AO systems, including Laser Guide Star
  (LGS) facility
• MCAO, MOAO, GLAO, MIRAO, ExAO
• Sensitivity D4 advantage for background-limited
  point sources with AO
• Wavelength range 0.31 - 28 microns (entire
  UV-mid-IR)
• Spatial resolution 0.007 at 1 micron, 0.014 at
  2 microns
• Instruments on large Nasmyth platforms, addressed
  by articulated tertiary
• Rapid switching between targets with different
  instruments (lt 10 min)
• (Rapid target acquisition time between targets lt
  5 min)

23
24
SAC Instrument Prioritization

• Desire to fund first-light instrument suite out
  of cost-capped construction budget
• Discovery space largest gains in broadest range
  of science in the near-IR (0.8-2.5
  microns)_at_diffraction limit
• IRIS IFUdiffraction limited imager
• IRMS multiplexed faint object spectroscopy in
  the near-IR -- leverages investment in facility
  MCAO system.
• Ability to perform guaranteed high-priority
  science we can think of now
• WFOS
• PFI very focused, but very powerful (GPI as a
  pathfinder...)
• HROS workhorse capability, strong science case
• Raw gains in sensitivity (D4) over existing or
  planned facilities, well defined science
• MIRES (mid-IR echelle)
• NIRES (near-IR echelle)
• WIRC (wider field diff. limited imager)

24
25
TMT First Decade Instrument/Capability Suite
25
26
TMT Early Light Instrument Suite
26
27
TMT Science and Flow-down

• Requirements for early light capabilities have
  been fine- tuned as a balance between unfettered
  science-driven desires and technical/fiscal
  realities (SAC/Project interactions have been
  crucial).
• We are proposing to build the most powerful suite
  of capabilities we can, through close interaction
  between science and engineering.
• Currently-envisioned capabilities address a huge
  range of questions we can formulate now (and
  complement other powerful facilities)
• The same capabilities will make new discoveries
  and will be the primary diagnostic tool for
  making sense of the discoveries made elsewhere.

27
28
IRIS Conceptual Design Team

• James Larkin (UCLA), PI, Lenslet IFS
• Anna Moore (Caltech), co-I, Slicer IFS
• Ryuji Suzuki, Masahiro Konishi, Tomonori Usuda
  (NAOJ), Imager
• Betsy Barton (UC Irvine), Project Scientist
• Science Team
• Mate Adamkovics(UCB), Aaron Barth(UCI), Josh
  Bloom(UCB), Pat Cote(HIA), Tim Davidge(HIA),
  Andrea Ghez(UCLA), Miwa Goto(MPIA), James
  Graham(UCB), Shri Kulkarni(Caltech), David
  Law(UCLA), Jessica Lu(UCLA),Hajime Sugai(Kyoto
  U), Jonathan Tan(UF), Shelley Wright(UCI)
• OIWFS (On Instrument Wavefront Sensor) Team (HIA
  Caltech)
• Led by David Loop, Anna Moore
• NSCU (NFIRAOS Science Calibration Unit) Team (U
  of Toronto)
• Led by Dae-Sik Moon

29
Motivation for IRIS

• Should be the most sensitive astronomical IR
  spectrograph ever built
• Unprecedented ability to investigate objects on
  small scales.
• 0.01 _at_ 5 AU 36 km (Jovians
  and moons)
• 5 pc 0.05 AU (Nearby stars companions)
• 100 pc 1 AU (Nearest star forming
  regions)
• 1 kpc 10 AU (Typical Galactic Objects)
• 8.5 kpc 85 AU (Galactic Center or
  Bulge)
• 1 Mpc 0.05 pc (Nearest galaxies)
• 20 Mpc 1 pc (Virgo Cluster)
• z0.5 0.07 kpc (galaxies at solar formation
  epoch)
• z1.0 0.09 kpc (disk evolution, drop in SFR)
• z2.5 0.09 kpc (QSO epoch, Ha in K band)
• z5.0 0.07 kpc (protogalaxies, QSOs,
  reionization)

M31 Bulge with 0.1 grid (Graham et al.)
Titan with an overlayed 0.05 grid (300 km)
(Macintosh et al.)
High redshift galaxy. Pixels are 0.04 scale
(0.35 kpc).Barczys et al.)
Keck AO images
30
WFOS/MOBIE Team

• Rebecca Bernstein (UCSC), PI
• Bruce Bigelow (UCSC), PM
• Chuck Steidel (Caltech), PS
• Science Team Bob Abraham(U Toronto), Jarle
  Brinchmann(Leiden), Judy Cohen(Caltech), Sandy
  Faber(UCSC), Raja Guhathakurta(UCSC), Jason
  Kalirai(UCSC), Gerry Lupino(UH), Jason
  Prochaska(UCSC), Connie Rockosi(UCSC), Alice
  Shapley(UCLA)
• Some flagship science cases, work horse
  capability
• High quality spectra of faint galaxies/AGN/stars
• IGM tomography
• Great follow-up and discovery potential -
  full wavelength coverage with spectral
  resolutions up to R 8000
• JWST, ALMA, etc., follow-up
• Sensitivity gt14 x current 8m telescopes

31
IR Multi-Slit Spectrometer(IRMS)

• IRMOS (deployable MOAO IFUs) deemed too
  risky/expensive for first light
• gt IRMS clone of Keck MOSFIRE, first step
  towards IRMOS
• Multi-slit NIR imaging spectro
• 46 slits,W 160 mas, L 2.5
• Deployed behind NFIRAOS
• 2 field
• 60mas pixels
• EE good (80 in K over 30)
• Spectral resolution up to 5000
• Full Y, J, H, K spectra (one at a time)
• Images entire 2 field

Slit width
Whole 120 field
32
IRMS Spectra

• Configurable Slit Unit originally developed for
  JWST (slits formed by opposing bars)
• Full Y, J, H, K spectra with R 5000 with 160mas
  (2 pix) slits in central 1/3 of field

33
Summary

• TMT will be a 30-m telescope with AO capabilities
  from the start
• 190 nm rms wavefront error over 10 arcsec
• First light 2018
• Very large and exciting science case
• 8 instruments planned for the first decade
• 3 instruments planned for first light
• IRIS (an AO NIR integral field spectrograph and
  imager)
• IRMS (an AO NIR multi object spectrometer (46
  slits)
• WFOS (a seeing-limited multiobject spectrometer
  with Rlt8000, and 50 arcmin2coverage)
• Many papers will elaborate on TMT AO in this
  conference