Lecture 21

1. Lecture 21 • Absolute flux calibration • Software • X-ray astrophysics – a very brief run through. • Swift & GRBs • 6.4 keV Fe line and the Kerr metric

2. Flux calibration • It is difficult to find x-ray sources which stay constant over time. • One can calibrate on the ground, but the ride to orbit shakes things up. • The Crab nebula (remnant of a supernova in AD 1054) is the best compromise. • The nebula itself is pretty stable but there is a pulsar (with x-ray pulses) in the centre. • Crab is also very bright – saturates modern, sensitive detectors like XMM. • ‘Crab units’ commonly used as a measure of x-ray flux.

3. Software packages • It seems common practice in the HE community to develop a new package for each new mission. • EG • ROSAT: EXSAS • XMM: SAS • Chandra: ciao • Not ideal either from software engineering principles, nor efficient use of resources. • Packages tend to be non-portable, non-generic • There’s also small incentive to document things properly! • Ciao is the best in my view. The USA tends to do software better than Europe – they spend more money on it for one thing. (But big is not always beautiful...) • SAS: some good generic FITS stuff there – but difficult to get at. • I’d like to do something about that!

4. Generic FITS stuff: • CFITSIO • Aging, user-unfriendly interface to FITS. Most more modern interfaces are built on top of it though. • FTOOLS • Very useful set of tools for doing everything you can think of to or with FITS-format data. • But too many tasks bundled into the one large, not-very-portable package. • DS9 • A very useful, compact, portable FITS image viewer.

5. X-ray astrophysics • Most sources appear to be compact – previously it was thought that there was diffuse emission both from the Milky Way and from much greater distances; however recent, more sensitive telescopes have resolved most of this into sources. • Accretion disks • X-ray binaries – small, nearby • AGN – large, far away • Compact  variable on short time scales. • Resolved (ie extended, non-compact) sources: • mostly clusters – x-rays from hot intergalactic gas.

6. Emission processes • Thermal – must have T ~ millions of kelvin. • Bremsstrahlung from optically thin gas, or • Black-body radiation from optically thick gas. • Synchrotron – ultra-relativistic electrons needed to get synchrotron at x-ray wavelengths. • Fluorescence (hence narrow spectral lines) • Inverse Compton scattering: Long-wavelength photon Short-λ photon e- Slow electron Fast electron e-

7. X-ray spectra: • Thermal: exponential decrease with E. • Synchrotron, inverse Compton: power-law decrease with E. • All show a fall-off at low E • This is due to photoelectric absorption by gas in the line of sight – mostly H. • Depends on the column densityNH in atoms cm-2. • Cutoff energy is (very roughly) ~ 3*10-9*NH0.4 keV.

8. Slowly moving bright object: Radiation is isotropic. ‘Normal’ Doppler shift. Object moving at relativistic speeds: Radiation is beamed. + sidewards Doppler. Relativistic jets – x-ray and radio aspects. v ~ 0 Shock fronts v << c Very dense central object – a source of strong gravity. v ~ c v ~ c Accretion disk – usually pretty hot. v ~ c v ~ 0 Shock fronts

9. Relativity? • The special theory of relativity: • effects of motion. • Beaming • Sidewards Doppler shift • The general theory of relativity: • effects of gravity. • Gravitational red shift • Time dilation

10. Jet radio emission Jets are (at least, we think) always symmetrical; but because of relativistic beaming, we only see the jet which is directed towards us (unless both go sideways). Cygnus A (this image spans a whole galaxy) Unbeamed synchrotron emission from slow electrons. Jet: synchrotron emission from relativistic electrons  beamed. Central source (prob. unresolved AD around a large BH) VLA 6 cm radio image. (Courtesy Dept of Astron, U Colorado)

11. Cygnus A in x-rays: Termination shocks Hot, tenuous galactic halo The ‘central engine’ – probably an accretion disk. Credit: NASA/UMD/A Wilson et al.

12. GRBs (Huge thanks to Paul O’Brien for many pictures.) • History: • 1963: VELA satellites launched – intended to monitor nuclear blasts. • 1972: VELA archival data on ‘non-Earth’ detections was examined.  serendipitous discovery of cosmic gamma ray bursts. • Indications that burst flux could vary on timescales < 1s  source must be small – must be time for a physical change to propagate across the detector. • 1991: BATSE detector of Compton/GRO. • Showed that sources were isotropic – NOT what was expected. • This means they are either very near by or very far away.

13. Non-GRB points to note: • Aitoff equal-area projection. • A new quantity: fluence. Milky Way coords

14. History continued • If bursts are close, one should eventually be able to detect a bias in the fainter tail. As time went on, the ‘nearby’ hypothesis came to seem less and less likely. • But, if the bursts are far enough away that they large-scale structure is smeared out, energy production must be gigantic! • 1996: BeppoSAX launched. • Sees about 1 GRB per day. • X-ray afterglows also seen. • First redshifts measured: average about 1. • 2004: Swift launched. • BAT: wide-angle gamma detector, to detect bursts; • XRT: narrow-angle, x-rays, more sensitive.

15. Swift instrumentation BAT: Wide-angle, detects burst, spacecraft steers to point XRT at the location. BAT XRT: high-res, narrow field, high sensitivity. XRT

16. How much energy in a GRB? E ergs Fluence = φ erg cm-2 D cm E = 4πD2φ. If average φ~10-6 erg cm-2, and average D~1028 cm, that gives E~1051 ergs. More energy than our Sun will emit in its entire 10 billion year lifetime! But beaming reduces this to E = ΩD2φ where Ω is the beam solid angle (which must be <= 4π).

17. What are GRBs? There seem to be 2 sorts: • Short, faint, hard bursts • Long, bright, soft bursts Frequency histogram (note the nice error bars!): Stellar collapse ? NS-NS ?

18. Long-burst GRBs: fireball-shock model Jet is so fast that the synchrotron is blue-shifted to gamma! External Shock Internal Shock The Flow decelerating into the surrounding medium Collision with surroundings Collisions in different parts of the flow X O GRB R Afterglow

19. Latest GRB history: • GRB 090423: redshift 8.2 – that’s huge. This breaks the record for the most distant object observed from Earth. • Only infrared afterglow seen for this GRB: all the visible light has been absorbed by the thin hydrogen haze between the galaxies. • Another recent (rather clever) discovery: • ‘Long’ GRBs seem to have very varied light curves. • But! There is a transform which brings them all into a common pattern.

20. Transformed light curves: Prompt Decay ОFlare О Hump Cheering news, because a common pattern implies common physics.

21. Fe lines in AGN – what this can tell us about GR • Fe: iron of course. There is a fluorescence line at 6.4 keV (in the rest frame). • AGN: Active Galactic Nucleus. • No-one has seen one close up – but almost certainly, these are accretion disks about a large black hole. • GR: here this doesn’t mean gamma ray, but General Relativity. • We expect to see GR effects near a giant black hole, because the gravitational force is so extreme.

22. (Tentative) cross-section through an AGN: To us Flare Jet Continuum background from halo Fe fluor. Flare additional continuum Accretion disk BH Accretion disk halo hot Very Innermost stable orbits – about 6rg for non-spinning BH; closer in for spinning BH. Jet

23. Fluorescence from a ‘cool reflector’. From A Fabian et al, PASP 112, 1145 (2000)

24. How ionization modifies this From A Fabian et al, PASP 112, 1145 (2000)

25. A Doppler-broadened spectral line – predictions of various theories: An accretion disk seen from above: Slow – small DS Profiles just for the two dashed annuli. Fast – large DS Fainter outside for whole disk Brighter inside From A Fabian et al, PASP 112, 1145 (2000)

26. Theory - spinning vs non-spinning BH Spinning (so-called Kerr-metric BH) Non-spinning (so-called Schwartzschild BH) From A Fabian et al, PASP 112, 1145 (2000)

27. Time-varying behaviour in MGC -6-30-15.This is a Seyfert I galaxy. From Tanaka et al (1995). 1997: time-averaged. 1994: time-averaged. 1994: in a deep minimum. 1997: at flare peak. From A Fabian et al, PASP 112, 1145 (2000)