1. AURA 2 Sep 2004

2. Context  National Academy’s Decadal Review recommended a ‘large synoptic survey telescope’ (LSST) 6m class aperture –dedicated wide-field optical imager –all-sky survey to enable multiple science goals –applications from solar system to cosmology –strategic emphasis in NEO threat  LSST is assumed to accept natural seeing for many applications the figure of merit is then simply the etendue: A  an alternative to a single 6m telescope is an array of smaller telescopes with the same total area A many pros and cons, the major advantages are: –single: no duplication of detectors, can go fainter faster –distributed: cheaper, faster to build, more flexible

3. AURA 2 Sep 2004 Pan-STARRS Design Philosophy  Given the following constraints: Construction time ~ 1 year per meter aperture Telescope cost rises faster than D 2 Pixel size limited to >10  m, desire 0.3” pixels requires a focal length of <8m Optical design for a large  becomes very expensive for fast f-ratios Costs of CCD detectors have been falling –(O)MEGACAMs: ~$8-10M for ~3x10 8 pixel or ~2-3c/pixel –Today it is possible to do a factor of 10 better  We believe it is cheaper and better to build an a survey instrument from an array of telescopes and detectors.

4. AURA 2 Sep 2004 Pan-STARRS in a Nutshell  Telescopes Four 1.8m R-C + corrector 7 square degree FOV Sited on Mauna Kea or Haleakala  Operation mode: Broad band optical imaging Four telescopes view the same field to detect transient or moving objects and build up a deep image of the sky  Partners: IfA: science, detectors, pipelines MHPCC: production DP, infrastructure SAIC: databases and mass storage systems MIT-Lincoln Lab: detectors  Detector and controllers 10 9 0.3” pixels per camera Image motion compensation 512 channel controller 2 second readout 4e - read-noise  Data-Processing System Multicolor summed images Difference images for detection of moving and variable objects Catalogs of static, moving, transient objects  Science: “Killer Asteroids” (PHAs) Huge range of other science topics – presently imagination limited

5. AURA 2 Sep 2004 Pu`u Poliahu UH 0.6-m UH 2.2-m

6. AURA 2 Sep 2004 UKIRT CFHT

7. AURA 2 Sep 2004 Science Overview  Time domain astronomy Transient objects Moving objects Variable objects  Static sky science Enabled by stacking repeated scans to form a collection of ultra-deep static sky images Observing Programs Science Programs  Very extensive overlaps between observational requirements of science programs!

8. AURA 2 Sep 2004 Pan-STARRS Surveys  Solar System (Ecliptic Plane) – used primarily to satisfy the observing requirements imposed by the PHO, NEO, MBA, KBO and other SS programs.  3  – used primarily to satisfy the observing requirements of the WL, LSN census, and EG object detection & classification programs; primary cadence drivers are the LSN census (and other proper motion studies)  Medium-Deep – the SNe, LSS, and the EG object detection & classification programs; primary cadence driver being SNe  Ultra-Deep – EG object detection & classification and, to some extent, SNe programs  Object Variability/Auxiliary – mostly user-defined supporting programs such as stellar variability and the search for extra-solar planets

9. AURA 2 Sep 2004 Design Reference Mission ModePSYAreaCad.wgrizy SS NEO 1.1d 0.2b 7000h/d/m 27.3 300 SS KBO 1.0d 0.2b 33 hdmy 26.3 60 Var. 0.8d 0.8b 1334 min 29.2 22000 28.6 7400 28.5 4400 24.9 4400 33 1.3d 2.5b 33 14d 25.9 30 25.6 30 25.4 60 23.9 20 22.3 30 Med. Deep 0.6d 0.9b 12004d 27.1 271 27.0 460 27.3 1200 25.0 1900 24.0 600 Ultra Deep 0.5d 0.7b 284d 29.1 10000 29.0 18000 28.0 6300 27.0 6700 26.0 26000 5-  limit (AB) Total int. (min)

10. AURA 2 Sep 2004 Science with Pan-STARRS  Moving Object Science NEO – Near Earth Object threat OSS/MBO – Main Belt and Other Solar System science KBO – Kuiper Belt Objects SOL – Solar Neighborhood (parallaxes and proper motions)  Static and Invariable Object Science WL – Weak Lensing LSS – Large Scale Structure LSB – Low Surface Brightness and dwarf galaxies SPH – Spheroid formation EGGS – Extragalactic and Galactic Stellar science  Transient and Variable Object Science AGN – Active Galactic Nuclei SNE – Supernovae GRB – Gamma Ray Bursts and afterglows EXO – Exoplanets (from occulation) YSO – Young Stellar Objects VAR – Variability Science (especially stars)  TGBN (Things that go Bump in the Night)

11. AURA 2 Sep 2004 Near Earth Asteroids Local Regional Global 10 8 Mton 10 5 10 3 10 2 100m1km10km 2x10 5 5x10 8 year Size Time Interval Energy

12. AURA 2 Sep 2004 Damage vs Size

13. AURA 2 Sep 2004 Risk Reduction vs Time (200m) (50m) (500m) (1000m) PS LINEAR LSST

14. AURA 2 Sep 2004 Inner Solar System Science  ~10 7 asteroids Families Orbit parameter space structure  ~10 4 NEOs Phase-space distribution Hazardous asteroids  Comets

15. AURA 2 Sep 2004 Outer Solar System Science  Kuiper Belt Objects Orbital distribution Formation and evolution  Trans-Neptunian Objects  Interlopers on hyperbolic orbits

16. AURA 2 Sep 2004 Stars and the Galaxy  Parallax survey Complete stellar census to 100pc  Proper motions Formation history  Other goals: Stellar variability Low mass stars Extra-solar planets

17. AURA 2 Sep 2004 Moving Objects  KBO – 20,000 KBOs over 10 years; all sky, unbiased. ~100 in binary pairs  OSS – many more asteroids and comets (~20x) 5x10 6 million main belt, 10 5 Jupiter Trojans, etc.  SOL – parallaxes to ~100pc in 10 years Best substellar IMF available (better than UKIDSS) 10-100x more brown dwarfs than SDSS or 2MASS  EGGS – proper motions of most stars in the Milky Way Accuracy of 2.5 km/s at 1kpc.

18. AURA 2 Sep 2004 Cosmology – Weak Lensing  Total mass power spectrum P(k) to large scales Test of inflation theory Evolution of P(k)  Higher order statistics Gravitational instability theory  Cluster mass function  Cosmology Cosmological parameters Geometric tests World model

19. AURA 2 Sep 2004 Static and Invariable Objects  WL – Weak lensing over 1000 sq deg. Large-scale structure of mass on large scales (wide area) and small scales (high density of objects) as a function of redshift, evolution of mass clustering. Mass profiles of galaxies  SPH, LSB, AGN – Evolution of galaxies Pan-STARRS will survey 4x the area of SDSS, will have the same photometric accuracy but 3-4 mag fainter, good sensitivity at 1um (y band). Reionization, metal formation, spheroid formation, AGN activity, galaxy merging, and cluster formation.

20. AURA 2 Sep 2004 Cosmology – Supernovae  Hubble diagram Dark energy equation of state w(z) Cosmological parameters  Supernova physics  Star formation history

21. AURA 2 Sep 2004 Transient and Variable Objects  SNE – 10,000’s of SNIa to z=1 Measure time (redshift) evolution of dark energy  AGN – Dropouts to z=7, variability identification Reionization, metals, spheroid formation, nature of radio sources, stellar disruptions, etc.  GRB – Optical counterparts (~100 per year) Possibly V~8 declining to V~20 in one day  EXO – Occultations of stars by planets Pan-STARRS is sensitive to Jupiters around sub- solar mass stars or Earths around brown dwarfs.  VAR – Stellar variability White dwarfs, binaries, Cepheids, Miras, RR Lyrae, microlensing, supergiants, etc, etc.

22. AURA 2 Sep 2004 TGBN  The Pan-STARRS survey is 10-20 times SDSS, Megacam survey, Vista, etc. in extent, but…  We are repeating it 30 – 500 times!  We will be the first to have extensive time domain information, designed with useful and interesting cadences, well controlled selection and systematics, and huge samples.  There is a high likelihood for unanticipated discoveries Unexpected variable objects Extremely rare objects Very large scale patterns

23. AURA 2 Sep 2004 Data Volume  Expect 700 images = 6 Tb per night raw; 3 Tb per night = 1 Pb per year reduced!  We have to be prepared not to “save the bits”  We must create a reliable enough pipeline that we tap all the science we want as the data flow through, and then throw the bits on the floor. (This has never been achieved before.)

24. AURA 2 Sep 2004 Image Processing Pipeline Phase 1 Detector Calibration (Calibration and Instrument Correction Processes) Phase 2 Map and Warp to Sky (Image Manipulation Processes) Phase 3 Create Sky Image (Image Combination Processes) Interface 1 Interface 2Interface 3Interface 4Interface 5 Data Storage Data Storage Data Storage Telescope Cameras Science Clients Phase 6 Science Client Interfaces TCS and Environment Monitoring Phase 4 Augmented Image Processing Data Storage Interface 7 Chip Level Telescope and System Level Interface 6 Phase 5 Science Client Product Generation Data Storage Data Storage System Level Image Capture Scheduler Phase 0 Internal Product Generation Internal Product Generation Mission Planning…pre-staging of each night’s scheduling and supporting data…TBD NB: specifics have changed!

25. AURA 2 Sep 2004 Confusing Issues  LSST should not be discussed as an either – or competitor to Pan-STARRS; Pan-STARRS will exist before LSST begins construction. Therefore: Astro-photo precursor survey will have been done, A robust data pipeline will have been shaken down, 50% of 300m PHAs will have been discovered, etc, etc.  What etendue is really needed? Etendue is A  those last two factors are important! SDSS at A  CFHT at 10, Suprime at 13 cannot approach LSST science because of limited  and/or  Pan-STARRS at A  is designed to have superb  and  and the software and scheduling to maintain LSST science. Pan-STARRS will improve on the present state of the art (SDSS, upcoming synoptic surveys) by at least an order of magnitude in science productivity.

26. AURA 2 Sep 2004 Design Reference Mission ModePSYAreaCad.wgrizy SS NEO 1.1d 0.2b 7000h/d/m 27.3 300 SS KBO 1.0d 0.2b 33 hdmy 26.3 60 Var. 0.8d 0.8b 1334 min 29.2 22000 28.6 7400 28.5 4400 24.9 4400 33 1.3d 2.5b 33 14d 25.9 30 25.6 30 25.4 60 23.9 20 22.3 30 Med. Deep 0.6d 0.9b 12004d 27.1 271 27.0 460 27.3 1200 25.0 1900 24.0 600 Ultra Deep 0.5d 0.7b 284d 29.1 10000 29.0 18000 28.0 6300 27.0 6700 26.0 26000 5-  limit (AB) Total int. (min)

27. AURA 2 Sep 2004 Final Data Products  Sky, the wallpaper: 10 Tpix x 6 colors x N versions  Sky, the movie: 10 Tpix x 6 colors x 50 epochs  Sky, the database: 2x10 10 objects (x 6 colors x 20-60 epochs) –Photometry to < 0.01 mag, astrometry to < 50 mas –Photometric redshifts of most of these objects 10 9 proper motions (complete over 3  ) 10 8 variable stars and AGN 10 7 asteroids (10 4 NEO/PHA) 10 7 transients (SN, GRB, etc.) 3x10 5 stars within 100 pc (with good parallax)