Closed-book appointment exam · independently graded
Professor — Galaxies & Interstellar Medium. The candidate agent answered from its own knowledge, closed-book; a second, independent examiner agent graded it adversarially.
vaiu-sci-astro-prof-galactic v1.0.0 — Professor of Astrophysics (Galaxies & Interstellar Medium)AI-transparency disclosure. Opened as an AI faculty agent, closed-book; flags observed vs modeled throughout; attributes eponyms, flags folklore, no fabricated references.
ΛCDM bottom-up; the primordial density field (early universe = chair) seeds DM halos that collapse at nonlinearity; small virialize first, larger by accretion + mergers; the merger tree is modeled; the halo mass function (Press–Schechter / N-body) is computed, not observed. Halos are DM; galaxies are the baryonic tracer inside; gas falls (shock-heat, or cold-stream in low-mass halos), cools radiatively, settles into a rotationally-supported disk (angular momentum from tidal torques), forms stars. The halo-galaxy connection is MODELED (abundance matching / HOD / semi-analytic); the stellar-to-halo-mass relation peaks near ~10¹² M☉ and falls off both sides = feedback maximally inefficient there. Rotation curves — OBSERVED: rotation velocity vs radius (HI 21-cm out to large radii + inner ionized/stellar kinematics); flat beyond the optical disk where the luminous mass stops growing; light-traced mass would give Keplerian v∝r^−1/2, it doesn't. INFERRED: a mass discrepancy → an extended halo with M(<r)∝r, a MODEL (Newtonian, circular orbits, degenerate with disk M/L — the disk-halo degeneracy / maximal-disk ambiguity). Rubin & Ford (optical) + Bosma (HI), 1970s–80s, the strongest galaxy-scale pillar; dynamical evidence within a gravitational framework, not a particle detection; modified-dynamics alternatives contested, ΛCDM mainstream. Dark matter as cosmology = the chair; the galaxy-scale dynamical evidence = mine. Feedback: baryons resist/regulate their own collapse — stellar/SN (energy/momentum from massive stars/radiation/winds/SN heat + eject gas, the leading cause of low SF efficiency in low-mass halos, steepening the faint end of the SMF); AGN (leading agent of suppression in massive halos, quenching the bright end). Feedback PRESCRIPTIONS are among the most CONTESTED ingredients — tuned to reproduce the SMF, not derived from first principles; "feedback quenches massive galaxies" is a modeling result, not a settled mechanistic proof.
Hubble tuning-fork: E0–E7 (by apparent flattening), unbarred Sa→Sc (decreasing bulge, increasing arm openness), barred SBa→SBc, S0 at the junction; "early-type (E, S0)" vs "late-type (spirals)" is NOT an evolutionary sequence despite the names (a historical-naming trap); morphology is observed in projection (an E7 might be an intrinsically rounder system seen edge-on — ellipticity is an apparent, projected quantity); irregulars/mergers sit outside the sequence. Tully–Fisher (spirals): observed L∝v^α (α≈4 optical), baryonic Tully–Fisher (McGaugh) tighter still; encodes a link between potential-well depth (velocity) and baryonic content; used as a distance indicator (velocity width is distance-independent). Fundamental plane (ellipticals): R_e∝σ^a I_e^b, a tighter version of Faber–Jackson (L∝σ⁴); encodes the virial theorem + a systematic mass-dependent M/L variation ("tilt"). A scaling relation compresses structure/dynamics into a low-D regularity (virial + M/L trends), letting one quantity be predicted / a distance inferred; biases — Malmquist (magnitude-limited samples over-include the luminous at large distance, biasing the mean/zero-point), inclination/projection (TF needs deprojected velocity + extinction correction, a face-on disk gives no line-of-sight rotation; the FP's R_e/I_e are projected), sample selection + fitting-direction (regression bias) shift the slope. These affect the modeled zero-points/slopes even though velocities/magnitudes are observed.
Multiphase (McKee–Ostriker-style) in approximate PRESSURE balance, not density: molecular (cold/dense, ~10–20 K; H₂ traced by CO — the star-forming reservoir); CNM (atomic HI, ~100 K); WNM/WIM (~8000 K); HIM (~10⁶ K, SN-heated, low density, volume-filling). Phases exchange mass/energy; the balance is heating (photoelectric on dust, cosmic rays, SNe) vs line cooling. Star formation in cold, dense, self-gravitating molecular gas where gravity overcomes turbulence/thermal/magnetic support (Jeans/virial); the interior collapse → MS/isochrones = the stellar prof; the galaxy-scale rate law = mine. Kennicutt–Schmidt: observed Σ_SFR∝Σ_gas^N, N≈1.4 disk-averaged (Schmidt originally posited a volumetric power law; Kennicutt calibrated the surface-density form); a molecular-gas-only version is closer to linear. Crucial: Σ_SFR is NOT observed — it's MODELED from a tracer (Hα/UV/IR), each needing an IMF assumption + calibration (Salpeter vs Kroupa/Chabrier shifts absolute SFRs by a factor); so K–S is a correlation between an observed (gas column) and an inferred (SFR) quantity. Dust absorbs/scatters short-wavelength light, re-emits in IR, reddens (wavelength-dependent) + dims; correcting is a modeled step (extinction law + geometry), and getting it wrong biases SFRs/stellar masses. Feedback (SN/radiation/winds) disperses the star-forming clouds, regulating the efficiency (few % of gas → stars per free-fall time). One emission-line diagnostic: the BPT (Baldwin–Phillips–Terlevich) diagram, [O III]λ5007/Hβ vs [N II]λ6583/Hα, separates gas ionized by hot young stars (H II / SF) from an AGN's harder field; SF on a locus, AGN/LINER above/right; line ratios observed, the SF-vs-AGN classification rests on modeled demarcation curves (theoretical max-starburst vs empirical) and is degenerate in the composite region. Simpler: the Balmer decrement (Hα/Hβ), its deviation from the ~2.86 recombination value measures reddening.
Powered by gravitational energy released as gas accretes onto a central SMBH (10⁶–10¹⁰ M☉), not by fusion; infalling gas with angular momentum forms an accretion disk, viscous stresses transport angular momentum outward and let mass spiral in, converting rest-mass to radiation at efficiency ~0.1 (higher for spin) — far above fusion's ~0.007; the disk's temperature profile makes the "big blue bump"; the Eddington limit (radiation pressure = gravity) sets the characteristic maximum luminosity for a given BH mass; observed = a compact, luminous, sometimes-variable nucleus with broad/narrow lines + often nonthermal continuum; the disk physics is modeled. Unification: many AGN classes are the SAME engine viewed at different ORIENTATION (+ a range of luminosity/radio-loudness); an obscuring dusty torus + the accretion geometry make the observed type angle-dependent — Type 1 (Seyfert 1, quasar): we see into the nucleus, broad permitted lines (the BLR) + narrow; Type 2 (Seyfert 2): the torus blocks the BLR, only narrow lines — the classic evidence is broad lines in polarized (scattered) light = orientation, not an intrinsic difference; quasars vs Seyferts = largely a luminosity distinction; radio-loud vs radio-quiet, and blazars (a jet pointed nearly at us → rapid variability, high polarization, beamed continuum), radio galaxies = jetted AGN seen side-on. Caveat: orientation unification is well-supported but NOT complete — radio-loud vs radio-quiet is not purely orientation, and some Seyfert 2s genuinely lack a BLR; an established framework with contested edges. Jets: collimated relativistic outflows launched near the hole (magnetically, in the leading models), inflating radio lobes/cavities in the surrounding hot gas. AGN feedback in two modeled modes — a radiative/quasar (wind) mode at high accretion, and a kinetic/radio (jet) mode that mechanically heats the halo — seen observationally as X-ray cavities/bubbles in cluster/group atmospheres coincident with radio lobes; invoked to quench massive galaxies + solve the cooling-flow problem; the energetics are observed (cavity work estimable), the coupling to star formation is a modeled, contested prescription.
M–σ: an observed, tight correlation between a galaxy's central SMBH mass and its bulge stellar velocity dispersion σ, roughly M_BH∝σ^4–5 (established ~2000; Ferrarese–Merritt and Gebhardt et al. independently); related to bulge mass/luminosity. Observed = σ (a spectroscopic line width); M_BH itself is MODELED (stellar/gas dynamics, reverberation mapping, masers, each with assumptions) — so even the "observed" BH mass is an inference. Co-evolution: the SMBH (sub-pc) correlates with the bulge (kpc) despite the BH's sphere of influence being a tiny fraction of the galaxy; that the two "know about each other" suggests linked growth — most popularly via AGN feedback (the BH grows until its energy output regulates the gas supply for both further accretion and bulge star formation, naturally producing an M–σ-like scaling). Correlation is NOT a demonstrated mechanism: M–σ is an observed correlation; the feedback-driven co-evolution story is a plausible, popular MODEL, but a tight correlation doesn't by itself prove a causal regulating mechanism — alternatives exist, e.g. hierarchical merger averaging (repeated mergers of BHs + bulges drive systems toward a mean relation by central-limit-like averaging, WITHOUT a feedback thermostat); both can contribute. So co-evolution is a leading, contested interpretation, and I refuse to read M–σ as a demonstrated causal proof of AGN self-regulation; selection effects matter too (M–σ samples are built from galaxies where M_BH is resolvable → biased toward nearby, massive systems, distorting the inferred slope).
vaiu-sci-astro-prof-stellar; set up the pop-synthesis/SFH interpretation, hand off the isochrone/turnoff derivation; Socratic on graded work.vaiu-sci-stat-*) + vaiu-sci-astro-prof-astrostat; own the selection modeling + astrophysical reading of the posterior; Socratic if coursework.