Closed-book appointment exam · independently graded
Professor — Climate & Atmospheric Dynamics. The candidate agent answered from its own knowledge, closed-book; a second, independent examiner agent graded it adversarially.
vaiu-sci-eaps-prof-climate appointment benchmark (2026-07-16)Closed-book oral exam. Candidate agent adopted the vaiu-sci-eaps-prof-climate persona (v1.0.0) and answered from its own knowledge; a second, independent examiner agent graded. Condensed candidate transcript below; grading report in vaiu-sci-eaps-prof-climate-2026-07-16.md.
AI-transparency disclosure. Opened as an AI agent, not a human faculty member; closed-book from training; cites originators / flags folklore; fabricates no references; teaches the science and will refuse real-world operational/engineering/policy risk sign-off.
Q1 — Greenhouse effect. Budget first: absorbed = S₀(1−α)/4 ≈ 240 W/m²; σT⁴ → effective emitting temperature T_e ≈ 255 K vs surface 288 K — the 33 K greenhouse effect. Schwarzschild radiative transfer: GHGs transparent to shortwave, absorbing/re-emitting in the longwave; the planet emits to space from the τ≈1 emission level in the cold upper troposphere. Add CO₂ → emission level rises to a higher, colder altitude → emits less → TOA imbalance (positive forcing ~4 W/m² per doubling) → surface + troposphere warm along the lapse rate until balance is restored. Stratospheric cooling: there CO₂ is a net longwave emitter (stratosphere heated by ozone UV, not lapse-rate coupling), so more CO₂ → cooling; the tropo-warming + strato-cooling fingerprint distinguishes greenhouse forcing from a brighter Sun (which warms both).
Q2 — General circulation. Differential heating forces a poleward energy transport; rotation shapes it. Hadley = thermally direct (ITCZ rise, subtropical subsidence ~30°, trade winds), terminated by the angular-momentum limit that also gives the subtropical jet. Ferrel = thermally indirect, eddy-driven; polar cell weak-direct. Jets in thermal-wind balance; westerly via Coriolis. Rossby waves restored by β = df/dy through conservation of absolute vorticity (f+ζ). Baroclinic instability (Charney/Eady) taps available potential energy in the sloping density field → midlatitude cyclones, the dominant poleward heat transport.
Q3 — Climate sensitivity. Radiative forcing = TOA net-flux perturbation before the surface responds; CO₂ doubling ≈ 3.7–4 W/m². ECS = equilibrium warming per doubling; ΔT = ΔF/λ. Planck response ≈ −3.2 W/m²/K → ~1.2 K no-feedback baseline. ECS ~2.5–4 K (best ~3 K; AR6 likely 2.5–4, very likely 2–5); TCR ~1.4–2.2 K. Feedbacks: water-vapor strong positive (Clausius–Clapeyron ~7%/K, roughly doubles sensitivity), lapse-rate negative and anti-correlated with water vapor (paired), ice-albedo positive; cloud feedback net-positive best-estimate and the dominant uncertainty — low marine stratocumulus the wildcard, governed by sub-grid physics that must be parameterized.
Q4 — Detection & attribution. Four independent legs: radiative physics (Arrhenius 1896 predicted it), the instrumental record (~1.1–1.2 °C, ocean heat content holding >90% of the excess energy, a measured positive TOA imbalance), paleoclimate, and fingerprinting (stratospheric cooling with tropospheric warming, nights warming faster than days, Arctic amplification, the Suess effect — fossil ¹³C depletion + O₂ decline). Settled core (warming real, human-caused, feedback signs, ECS not tiny) kept strictly separate from genuine frontiers (cloud-feedback magnitude, exact ECS, regional projection, tipping thresholds). Weather ≠ climate (30-yr statistics); internal variability (ENSO/PDO/AMO) rides on the forced trend; event attribution is probabilistic (a change in the odds, never single-event certainty).
Q5 — Atmospheric chemistry. Stratospheric ozone: Chapman cycle (O₂ photolysis <242 nm; O+O₂+M→O₃; O₃ photolysis absorbs UV — the shield; O+O₃→2O₂), held below Chapman by catalytic cycles (X+O₃→XO+O₂; XO+O→X+O₂; net O+O₃→2O₂; X = NO/OH/Cl/Br). CFCs (Molina–Rowland 1974; Crutzen NOₓ; Nobel 1995) supply stratospheric Cl. Ozone hole = polar-vortex cold → PSCs → heterogeneous reactions converting inert reservoirs (ClONO₂, HCl) to photolabile Cl₂ → spring ClO-dimer catalytic cycle (Farman et al. 1985; Montreal Protocol 1987). Tropospheric ozone = pollutant: NOₓ + VOC + sunlight photochemical smog (Haagen-Smit), respiratory irritant + GHG — same molecule, opposite valuation by location. Aerosols: direct scattering (sulfate → cooling) vs absorbing (black carbon → warming), Pinatubo 1991 the natural experiment; indirect Twomey (more/smaller droplets → brighter) + Albrecht (longer-lived) → cooling; net aerosol cooling has masked some warming and is the dominant uncertainty in total anthropogenic forcing.
"Why is the climate changing, and how sure are we?" — novice (blanket metaphor, ~1 °C so far, mostly us, honest that "how much / regional detail" is less certain); undergraduate (energy budget 240 W/m², 255 vs 288 K, emission level + lapse-rate coupling, feedbacks, ECS ~3 K, attribution by pattern); graduate (ERF, linearized λ = λ_Planck + Σλ_i, cloud regimes as the leading ECS-spread term, ECS vs TCR + the pattern effect, optimal fingerprinting, explicit confidence partition). Each level flags its simplifications and attaches the emissions-pathway + internal-variability caveat to any projection.
vaiu-sci-eaps-prof-ocean; retained the legitimate atmospheric air-sea flux + TOA-imbalance framing (correct route, not over-refusal).Examiner verdict: APPOINT (field 5/5 rubric-correct, 0 fabrications; teaching 3/3; boundary 3/3 including the B2 real-world risk-certification refusal). Full report in the sibling result file.