1. NASA’s Gravitational-Wave Mission Concept Study Robin Stebbins, Study Scientist Ninth LISA Symposium Paris, 22 May 2012

2. Outline Goals Elements of the study Context of the study Responses to the Request-For-Information (RFI) Science performance analysis Assessment of architectures Risk Cost 2 This document contains no ITAR-controlled information and is suitable for public release.

3. Goals of the Study Develop mission concepts that will accomplish some or all of the LISA science objectives at lower cost points. Explore how mission architecture choices impact science, cost and risk. Big Questions Are there concepts at $300M, $600M or $1B? What is the lowest cost GW mission? Is there a game-changing technology that hasn't been adequately considered? 3 This document contains no ITAR-controlled information and is suitable for public release.

4. Elements of the Study Request for Information (RFI) – due Nov. 10 th. Core Team – ~25 GSFC, JPL & university scientists and engineers critically reviewing RFI responses Science task force – ~15 volunteer scientists evaluating science performance of concepts Community Science Team (CST) – 10 scientists, Rai Weiss, Ned Wright co-chairs Public workshop – December 20-21 st Concurrent engineering studies by JPL’s Team-X in March and April Final Report to NASA Headquarters – July 6 th Presentation to the Committee on Astronomy and Astrophysics (CAA) of the National Research Council (NRC) This document contains no ITAR-controlled information and is suitable for public release. 4

5. Context of the Study – A Brief History of LISA 1974 - A dinner conversation: Weiss, Bender, Misner and Pound 1985 – LAGOS Concept (Faller, Bender, Hall, Hils and Vincent) 1993 – LISAG - ESA M3 study: six S/C LISA & Sagittarius 1997 - JPL Team-X Study: 3 S/C LISA 2001-2015 - LISA Pathfinder and ST-7 DRS 2001 – NASA/ESA project began 2003 – TRIP Review 2005 – GSFC AETD Review 2007 – NRC BEPAC Review 2009 – Astro2010 Review 2011 – NASA/ESA partnership ended 2011 – Next Generation Gravitational-Wave Observatory (NGO) started 2012 – ESA L1 downselect 5 This document contains no ITAR-controlled information and is suitable for public release.

6. Context of the Study – Activities in Europe LISA Pathfinder Demonstration of space-based GW technology, in late stages of I&T 2014 launch Technology development Inertial sensor electronics, charge control Optical system Laser system Pointing and point-ahead mechanisms NGO Highly developed concept with extensive science case and technical detail 6 This document contains no ITAR-controlled information and is suitable for public release.

7. Context of the Study in the U.S. Next major mission in Astrophysics starts after 2018. The Astrophysics Division anticipates that a “probe- class” mission could be started ~2017. The Division will not commit to a ‘large’ mission until after Astro2020. ‘Commit’ means the Confirmation Review at the end of Phase B. A partnership with ESA seems highly likely. That would require: Rebuilding a partnership Reliably coordinating two agencies’ programs 7

8. RFI Responses 8

9. 17 responses total 12 for mission concepts, several with options 3 for instrument concepts 2 for technologies Four natural groups No-drag-free concepts (2) Geocentric orbits (4) LISA-like (5) Other (2) 9

10. What constitutes “LISA-like?” Drag-free control Free-falling test mass Precision stationkeeping Continuous laser ranging Heliocentric orbits Constellation in stable equilateral triangle No orbital maintenance Million-kilometer long arms Laser frequency noise subtraction (TDI) Emulate Michelson’s white-light fringe condition through post-processing 10

11. No-Drag-Free Concepts 11

12. No-Drag-Free Concepts Rely on either very long arms (50X LISA) or geometry (100X reduction) to compensate for using the spacecraft as the test mass. Disturbances are solar radiation pressure variability, solar wind, interplanetary magnetic field Measure, model and correct for spurious forces (10 2 - 10 4 X) Displacement noise from motions of the spacecraft CG, owing to, say, thermoelastic effects Concerns about measuring solar wind and modeling/testing other disturbance (e.g., Pioneer effect) 12

13. Geocentric Concepts 13

14. Geocentric Concepts Noise concerns Thermal environment: moving sub-Sun point, eclipses Sun in the telescope Varying Earth albedo Geosynchronous may have interesting modulation properties. (McWilliams’ talk Thursday afternoon) LAGRANGE/Conklin described by Buchman Tuesday afternoon. A big cost question: can you do this for a factor of 4 less by employing nanosat technology, lower reliability standards, standard bus, a different way of doing business, … a different business model? 14

15. LISA-like Concepts 15

16. LISA-like Concepts How far can the LISA architecture be descoped? No technical or performance issues Science performance falls off much faster than cost Found the bottom! See Jeff Livas’s talk Tuesday afternoon in LISA-NGO Technology session. 16

17. Other Concepts 17

18. Other Concepts The superconductor idea doesn’t work. Atom Interferometry Atoms clouds as test masses Atom interferometer as a phasemeter See John Baker’s talk Thursday afternoon in Other Experiments and Alternative Design session InSpRL Most aggressive design concept Invoked ‘superclocks’ and resonance Seems to require a few orders of magnitude improvement in several key performance parameters Lacks enough definition to evaluate Yu concept doesn’t promise to be cheaper. Digital Interferometry is interesting. 18

19. Science Performance Analysis 19

20. Science Performance 20 Volume of the Universe explored Detection numbers for source populations (Massive BHs, EMRIs, Galactic Binaries) Discovery space Parameter resolution All work done by Neil Cornish and the Science Task Force. See Cornish talk, Friday morning.

21. Sensitivity Curves – All 15 Concepts 21

22. Massive Black Hole Horizons 22

23. Massive Black Hole Horizons – No-Drag-Free 23

24. Massive Black Hole Horizons – Geocentric 24

25. Massive Black Hole Horizons – Geosynchronous 25

26. Massive Black Hole Horizons – LISA-Like 26

27. Detection Rates – Large Seed Models (/yr) 27

28. Detection Rates – Small Seed Models (/yr) 28

29. EMRI Horizons 29

30. EMRI Detections 30 10 M ⊙ compact object, eccentricity 0.5 at 2 yrs to plunge, spin 0.5 central BH, SNR=15

31. WD-WD Detection Numbers 31

32. Parameter Estimation – LISA-like Concepts 32 Similar detection numbers, but each descope x 3-10 loss in resolution

33. Architecture Choices 33

34. Architecture Choices – Mission Design Heliocentric – fixed, drift-away, in-line, L2/leading/trailing, 1 AU Geocentric – OMEGA, geosync, L3/L4/L5, LEO Compare delta-v, constellation stability, propellant, thermal, modulation of science signal, comm 34

35. Architecture Choices – Inertial Reference Proof mass – cubical, parallelepiped or spherical free-falling, or torsion pendulum Spacecraft center-of-gravity (aka no-drag-free) with modeled corrections Atom interferometry - atoms as proof masses, atoms as secondary inertial reference Payload as separated spacecraft 35

36. Architecture Choices – Measurement Strategies Laser interferometry with laser heterodyne phase comparison – free-space or digital interferometry Laser interferometry with atom interferometer phase comparison Laser and clock frequency noise correction – 3 spacecraft & TDI, or very much better phase reference (AI) 36

37. Implementation Strategies 37

38. Implementation Strategies 38 ParameterSGO MidLAGRANGEOMEGA Mass Margin 53% Payload mass (kg), power (W) CBE 216.5 kg, 233 W99.7 kg, 99.3 W Option 1: 64.3 kg, 80W; Option 2: 55 kg, 54W Mass rack-up Science-craft type 1 Science-craft type 2 Propulsion Module type 1 + Prop Propulsion module type 2 + Prop LV Adapter Launch Mass Wet 717 kg (3) 661 + 139 (3) ? 4553 kg 531 kg (2) 586 kg (1) 224 + 174 (2) 591 + 114 (1) 32 kg 3182 kg 147 kg (6) 374 + 465.5 (1) 28 kg 2347 kg Launch Vehicle Atlas V 551; 6075 kg Atlas V 511; 3285 kg Falcon 9 Block 2; 2490 kg

39. Risk 39 SGO-Mid/HighLAGRANGEOMEGA These are a combination of Team-X and Core Team risks. Risk rises rapidly with modest (<10%) cost reductions. This assessment is not complete.

40. Cost Team-X is very conservative. Cost estimates range from $1.2B to 2.1B. Per science year costs SGO-hi $450M/yr SGO-mid/Lagrange ~$800-900/yr Omega ~$1,300M/yr Important cost drivers Non-recurring costs (NRE) and recurring costs (RE) are important. Design validation Serial vs parallel construction of multiple units (~$150M/yr) 40

41. Summary The CST prefers SGO-Mid (3 arms, LISA-like, 1 Mkm, drift- away). Big Questions We found no concepts at $300M, $600M or $1B. The lowest cost GW mission is ~$1.4B (±0.2). We found no game-changing technology that hasn't been adequately considered. Heliocentric is a better choice than geocentric. Three dual-string spacecraft appear to be more robust than six single-string spacecraft. No-drag-free achieves only modest savings while incurring substantial risk. [Cost model is uncertain.] Three arms has lower risk and mediating cost factors relative to two arms. 41

42. Backup Slides 42

43. Feedstock Whitepapers (17x~15 pages = 235) Workshop Presentations (~20 x 30 charts = 600) Core Team Work (~200 pages) Team-X input Presentations (4 x ~60 charts = 240) Master Equipment Lists Functional Interface Diagrams CAD files Orbit analyses Team-X output Viewgraphs (~3 x 280 = 840) Team-X reports (~3 x 10-20 = 45) CST Work (~50 pages) Total: north of 2230 pages 43

44. Mission Design Review 1/2 44 FeatureSGO-MidLagrangeOmega 1. Trajectory Phase  V [174, 153, 200] m/s Stack ~ 120 m/s to L2 [SC1, 3]: [460, 300]  [206, 328, 450] + 4 m/s vs. 3  210 m/s if 3 PMs Significance: Prop module size(s), Total mass, Launch vehicle 2. Trajectory Phase  t 17 months 27 months  12 months (vs. ~ 7 ) Significance: Cost/complexity of trajectory phase operations (FDF & Ops) 3. Lunar Flybys Used No Yes  No Significance: Cost/complexity/risk of trajectory phase operations (FDF & Ops) 4. Mission Phase  t 2 yr / extendable2 yr / not extendable1 yr / extendable Significance: Cost of science operations, Amount of science, Constellation Stability 5. Const. Stability  L/L, , , (  ||,  + ) ±0.007, ±0.6 , ±1.5 Mhz, ±(0.008, 1.0)  rad ±0.1, ±0.12 , ±94 Mhz, ±(0.8, 0.32)  rad  ±0.025, ±2.2 , ±60 Mhz, ±(0.17, 0.15)  rad  Significance: Cost of additional mechanisms and electronics 6. Mission Phase  V No Yes (SC2)  No Significance: Cost/sophistication of mN-thruster system (~ 10 m/s/yr)

45. Mission Design Review 2/2 45 FeatureSGO-MidLagrangeOmega 7. Distance to Earth / HGA, ISC req? 24 to 55  10 6 km / HGA [21, 1.5, 21]  10 6 km HGA/LGA, ISC/LGA 0.6  10 6 km LGA Significance: Cost/complexity of communications; ISC = inter-spacecraft comm. 8. GeoEcliptic OrbitNo Yes  Significance: (a) Sun direction variation (thermal stability) (b) Sun in telescope aperture (thermal, optical interference) (c) Earth eclipses (thermal, science interruptions) FeatureSGO-MidLagrangeOmega 9. Launch Vehicle C3+0.27 (km/s) 2 -0.3 (km/s) 2 -0.05 (km/s) 2 Significance: Launch vehicle selection 10. Single Prop Option No (  ?) Yes (but not necessary) Significance: Input to possible trade for single prop module cost savings (?) 11. “FDF” Ops Cost $ 18 M $ 27 M  $ 23 M (? if 3 PMs ) Cost Drivers: Trajectory and mission phase durations, trajectory complexity