1. IAC-04-IAA-1.1.1.06 SEARCHING FOR DYSON SPHERES WITH PLANCK SPECTRUM FITS TO IRAS Dick Carrigan Fermilab Infrared Processing and Analysis Center, Caltech/JPL. IPAC is NASA's Infrared Astrophysics Data Center.

2. International Astronautical Congress Vancouver D. Carrigan - Fermilab Oct. 4 – Oct. 8, 2004 2 The radio SETI paradigm SETI radio beacon (acquisition signal) –but why? Material and electromagnetic ET artifacts like a Dyson Sphere don’t require reason to communicate A substantial fraction of sun-like stars out to several hundred light years have been monitored for ETI with radio SETI. Credit: Allen Telescope

3. International Astronautical Congress Vancouver D. Carrigan - Fermilab Oct. 4 – Oct. 8, 2004 3 Cosmic archaeology … and Dyson Spheres Nuclear waste disposal Salting with unusual atomic lines – Technetium-Drake and Shklovskii – 10 5 tons. G. Lemarchand, SETIQuest, Volume 1, Number 1, p. 3. On the web at http://www.coseti.org/lemarch1.htmSETIQuest, Volume 1, Number 1, p. 3. Kardashev civilizations: I-planet, II-star, III-galaxy Annis-JBIS 52, 33 (1999)-elliptical gal log(v dispersion)+sur bright (mag) vs log radius outlier line 1.5 mag or 75% of energy Salting star to change Hertzsprung-Russell curve- unlikely may require too much material Star lifting (D. Criswell)

4. Rigid Dyson sphere is unstable instead swarm of smaller pieces Searches-problem fixing distance Signature infrared stellar luminosity (distance problem) pure Planck no star for pure DS Types of Dyson Spheres - …pure, partial, rings, Dyson spheres Dyson Sphere 4 mm thick Venus Sun R=1.5*10 8 km Infrared radiation Based on www.daviddarling.info/encyclopedia/D/Dysonsp.html Types pure – star completely obscured partial Energy to assemble 800 solar years to take Jupiter apart Sagan and Walker, Astrophysical Journal 144(3), 1216 (1966) search feasible even with sixties technology but that the possible confusion with natural signatures could require searches for other artifacts of intelligence

5. International Astronautical Congress Vancouver D. Carrigan - Fermilab Oct. 4 – Oct. 8, 2004 5 Dyson Sphere surrogates Mira (Omicron Ceti) in visible (Hubble image) Miras variables, short-lived, circumstellar dust Sum of many Planck spectra Also C stars Protostars forming in Orion dust cloud (IRAS image) Brown dwarfs but temperature is typically higher absolute luminosity is lower ←Planetary nebula from IRAS dumbbell M22. IRAS 06176-1036 “Red Rectangle” →

6. Earlier searches Jugaku and Nishimura, Bioastronomy 2002: Life Among the Stars, Norris and Stootman, eds, IAU Symposium, 213, 437 (2002) and earlier use the 2.2 μm K band as an indicator of the photospheric radiation of a star hosting a partial Dyson Sphere and then look for an infrared excess in the IRAS infrared satellite 12 μm band 1 mag difference for 1% sphere. See less than 0.3 mag for 384 stars inside 25 pc. Slysh, in The Search for Extraterrestrial Life: Recent Developments, Papagiannis (Ed) 1985. Timofeev, Kardashev, and Promyslov, AAJournal, 46, 655 (2000) [TKP] Four band IRAS Planck fit. Several candidates, limited sky. 100, 300 °K. 98 stars Conroy and Werthimer, preprint (2003) Jugaku technique to older stars. 1000 nearby older stars from Wright and Marcy Older stars eliminate thick dust clouds around young stars Correlate with the K band near-infrared ground based data from 2MASS 33 candidates in the 12 μm IRAS band with 3 σ excesses from mean. Globus, Backman, and Witteborn, preprint (2003) look for a temperature/luminosity anomaly due to the fact that the luminosity of a star surrounded by a partial DS would be lowered compared to naked star with same T

7. International Astronautical Congress Vancouver D. Carrigan - Fermilab Oct. 4 – Oct. 8, 2004 7 IRAS Picture from Infrared Processing and Analysis Center, Caltech/JPL. IPAC is NASA's Infrared Astrophysics Data Center. Only available all-sky survey at 12 μm 12, 25, 60, 100 μm micron filters A main purpose – dust, mirror only 0.6 m cosmic cirrus problems in 100, 60 μm Performance sensitivity – 0.5 Jy 12 – 60 μm, 1 Jy for 100 μm 250 K point sources angular resolution – O(1') positional – 2 to 6" in-scan, 8 – 16" cross Requirement for Dyson Sphere search all sky – useful 100 < T < 600 °K 2MASS much more sensitive, 500 M point sources IRAS 12 μm must be at least 10 Jy to register in the 2MASS 2.17 μm filter

8. International Astronautical Congress Vancouver D. Carrigan - Fermilab Oct. 4 – Oct. 8, 2004 8 Analysis Infrared cirrus significant presence of emission in the 100 μm band on a wide range of angular scales from so-called infrared cirrus due to interstellar dust often well above Planck do not use 100 μm for fit (Slysh and TKP used all four filters) Flux quality factors, FQUAL(i) do not use if only an upper limit leaves 19572 sources limits upper temperature range e. g. missing 60 μm looks like high temperature DS Temperature range limited to 150 to 500 ° K

9. International Astronautical Congress Vancouver D. Carrigan - Fermilab Oct. 4 – Oct. 8, 2004 9 Planck fitting (IRAS 06176-1036 “Red Rectangle”) Solid red is trial fit to three points Dotted is final fit F is IRAS flux, K is Planck color correction, P is Planck dist., sigma is weight, and a is fit para

10. International Astronautical Congress Vancouver D. Carrigan - Fermilab Oct. 4 – Oct. 8, 2004 10 Color color fitting Left is 12/25, rt 25/60 Fit with arbitrary polynomial Diamonds are extragalactic Dots - no catalog entry, pure DS Dyson Spheres along BB line Follow Pottasch et al. AA 205, 248 (88), Fig. 1 Each pair gives a black body temperature line shows equal pairs, dots BB temp 500 400 300 200

11. International Astronautical Congress Vancouver D. Carrigan - Fermilab Oct. 4 – Oct. 8, 2004 11 Relation of color-color fits to Planck fits: no star ↓blackbody The lowest LSQ rises quickly as dT/T moves away from 0 By dT/T = 0.2 LSQ almost half of the maximum LSQ at dT/T = 0 may be possible to rule out LSQ values greater than 1-2E-7 There is a direct relation between dT/T and LSQ

12. International Astronautical Congress Vancouver D. Carrigan - Fermilab Oct. 4 – Oct. 8, 2004 12 dT/T distribution from color color fitting No obvious peaking at dT/T = 0 where pure Dyson Sphere should be 370 sources with no catalog entry between -0.1 < dT/T < 0.1 so that 1 out of every 600 IRAS sources in interval However distribution statistically flat in the region of dT/T = 0 A 3 σ peak in one bin might require about 25 sources or one in 10,000 of the IRAS sources. ↓blackbody line

13. International Astronautical Congress Vancouver D. Carrigan - Fermilab Oct. 4 – Oct. 8, 2004 13 Future plans Determining the absolute luminosity need distance to source possible ways to determine distance clusters like Pleiades multiple star system center of galaxy at 8 kpc partial of Jugaku or Dyson Sphere – use source distance of companion The Pleadies at 125 pc contains about 1200 objects subtends a field of roughly four square degrees or approximately 10 -4 of the sky contain about 25 members of a randomly distributed sample of 250K sources Thus association with a nearby cluster will be a useful tool for less than 0.1% of the objects in IRAS sample Small dT/T Dyson Sphere candidates individual cases need to be matched with other information available on the source Following TKP one can look for further information from mm wave measurements, 2MASS, or even additional measurements using SIRTF

14. 14 Summary For pure Dyson Spheres do need distance to get luminosity. Clusters, galactic center, Virgo? Artifacts like pyramids, Dyson Spheres and Kardashev civilizations are “natural” and don’t require purposeful signals Dyson Sphere Ir rad IRAS good-whole sky, problems-angular resolution. Still-best compromise. At 3 σ in one bin is 1 in 10,000 of IRAS Can get good black body fits and do find candidates but there are surrogates like the planetary nebula “Red Rectangle”

15. International Astronautical Congress Vancouver D. Carrigan - Fermilab Oct. 4 – Oct. 8, 2004 15 Questions?

16. International Astronautical Congress Vancouver D. Carrigan - Fermilab Oct. 4 – Oct. 8, 2004 16