1. SETI: Search Strategies & Current Plans Jim Cordes 23 September 2002 Motivation for searching we’re here life expected to be common (especially microbial life) technology common? Yes  N>>1 No  N=1 Search issues Cross section of SETI programs so far Future SETI Editorial comments

5. Parkes MB Feeds Arecibo Multibeam Surveys

6. Current SKA Concepts China KARST Canadian aerostat US Large N Australian Luneburg Lenses Dutch fixed planar array (cf. Allen Telescope Array, Extended VLA) (cf. LOFAR = Low Freqency Array)

7. Expand The Galactic Exploration ATA Phoenix SKA

8. Why search? Assessing the Odds The astrophysical case: p(habitable planets | Galaxy) The biological case: p(life | habitable planets) Complexity: p(technology | life) p(extroversion | technology)

9. Related Issues Copernican principle – we’re mediocre & there must be lots more like us Anthropic principle – the universe necessarily has properties that allow complex (but mediocre) beings like ourselves to have evolved. Fermi Paradox – given CP + AP, Where is everybody?

10. Additional SETI Issues Large N  optimism about evolutionary trends leading to technological life, its longevity, and perhaps about Galactic colonization Counterpoints: –What took hominids so long to evolve on Earth? –‘Rare Earth’ arguments (Ward & Brownlee) Our preconceptions about N have a strong influence on –how luminous ET transmissions must be for detection –beaming of ET transmissions (toward us?!) N determines how far we must look in the Galaxy How far we look determines the role of propagation effects from ISM plasma (radio) or grains (IR/optical)

11. SETI Conundrums Deliberate transmissions Leakage transmissions RadioOptical / IR NarrowbandPulsed Large NSmall N High LuminosityLow Luminosity

12. What do we look for? Reciprocity: what do we radiate? Radio typical: detectable to ~ few pc strongest: planetary radar ~1 kpc Optical/IR typical: nil pulsed IR lasers: ~ few x 10 pc  -rays 1 Mton: ~ 1 AU with CGRO

13. Rationale for Radio SETI No Galactic absorption No background from host stars Maximum S/N in microwave band (1-10 GHz) Magic frequency arguments e.g. 1.42 GHz H 1.67 GHz OH  x 1.42 GHzetc. Narrowband signals << thermal Doppler widths of natural, astrophysical sources Propagation effects (dispersion, scintillation, pulse broadening from ISM) are important

14. Notable Radio SETI Programs Ozma 1960 (Frank Drake) targeted (two nearby stars) Serendip I-IV (piggyback surveys at Green Bank and Arecibo) blind surveys (1970s – present) NASA targeted survey + sky survey (1992-2001, but cancelled in 1993) targeted: ~1000 nearest G-type stars, single, age > 3 Gyr sky survey: full sky, 1- 10 GHz Phoenix = privately funded version of NASA targeted survey (SETI Institute; uses Arecibo)

15. Notable Radio SETI Programs META blind survey (Harvard) 1985-1993 magic frequencies (1.42 GHz & x2) ~ 10 6 channels META II Argentina BETAblind survey (Harvard), present L band (H, OH), ~ 10 9 channels Serendip IV blind survey, (Berkeley), present L band, Arecibo, ~ 3x10 8 channels SETI@Home blind survey, ongoing, L band, baseband sampled data (Arecibo), software data reductionSETI@Home

17. Optimizing radio SETI against background noise

18. Anticipated radio ET Signals (by ‘strong SETI’ proponents) Narrowband (~ 1 Hz) Weakly modulated (~ 1 bit/s) Drifts in frequency (orbital + planetary motion) df/dt ~ 10 to 100 Hz/hour (some argue that deliberate transmissions to us would be Doppler corrected) Pattern recognition algorithms: search for narrowband, drifting features in the frequency-time plane (dechirping algorithms) Search space: (B/  )(T/  t)N sky > 10 13 trials  need very high threshold (e.g. 30  ) to achieve small false-alarm rate

20. Spectra from the OH masers in W49 Power spectra calculated from baseband sampled data from Arecibo; Signal statistics = exponential

22. Real-world effects Terrestrial & spacecraft radio frequency interference (RFI): diverse, mimics anticipated signals (our RFI = their ETI signal and vice versa) Interstellar scintillation causes deep fading and occasional amplification

25. Electron density irregularities exist on scales from ~ 100’s km to ~ pc as approximately a power-law spectrum (~ Kolmogorov) Pulsar velocities >> ISM, observer velocities 500 km/s average (100 to 1700 km/s) Isoplanatic angle ~ 10 -6 arc sec  AGNs don’t show DISS, pulsars do Expect ETI sources to show DISS  Deep fading & amplification (100% modulation)  longer time scales than pulsars (lower velocities)

26. INTERSTELLAR DISPERSION DM =  0 D ds n e (s) Known for ~1200 pulsars DM ~ 2 to 1100 pc cm -3 Variable at ~10 -3 pc cm -3 Variations with d,l,b show obvious Galactic structure

27. Dynamic spectrum of pulsar scintillation

28. Narrowband signals will show deep modulation with exponential statistics

32. Optimizing a search: better to split total time per target into ~4 intervals so that ISS is uncorrelated between them

35. Rationale for Optical/IR SETI Pulsed lasers distinguishable from host star with reasonable power (nanosecond pulses) Optical/IR not susceptible to ISM plasma propagation effects … But interstellar absorption and scattering from grains important for optical and near IR (scattering  smearing of pulse)

36. Laser power Petawatt (10 15 watts) pulse lasers exist for laser fusion, are sufficient to produce detectable pulses from systems on planets around G-type host stars. For ns pulses, a 1-m telescope + photomultiplier is sufficient to detect sources out to ~ 30 pc. Programs at Berkeley, Harvard, amateur.

37. I. Arecibo Galactic-Plane Survey |b| < 5 deg, 32 deg < l < 80 deg 1.23-1.53 GHz bandwidth = 300 MHz digital backends (<0.3 MHz channels) –Correlator based e.g. 7 x (2 x WAPP)? (200 MHz) –FPGA-FFT or Polyphase filter approach? (300 MHz) ~300 s integrations, 3000 hours total Can see 2.5 to 5 times further than Parkes MB –period dependent –from AO sensitivity + narrower channels (larger DM) Expect ~1000 new pulsars

38. Surveys with Parkes, Arecibo & GBT. Simulated & actual Yield ~ 1000 pulsars.

39. SKA advantages: Multibeaming, multiple sites One station of many in SKA

40. Comments Observational phase space is very incompletely covered to date [, F,  t, , d /dt, transients, etc.] many of the radio sources in large scale surveys remain unidentified (though many are likely to be AGNs, pulsars, microquasars, HII regions and flare stars)  empirical conclusions about N not yet possible SETI strategies that strongly leverage notions about the motivations of ETI are not robust in their ability to constrain N An economical approach is to design telescopes & surveys for astrophysical purposes & conduct SETI as a subset or spinoff of the sky coverage  requires SETI specific digital backend systems that exploit Moore’s law. [, F,  t, , d /dt, transients, etc.]  empirical conclusions about N not yet possible