v1.3.0

sdo.aia

New: an AIA temperature response computed from CHIANTI 11 (#28). temperature_response() takes emissivity="SolarSoft" | "CHIANTI 11".

• Default: "SolarSoft", the V10 table, as before.
• Same inputs as V10: the new table is computed from CHIANTI 11.0.2's lines, continua and ionization equilibrium. Everything else matches V10: the V9 effective areas, Feldman (1992) extended coronal abundances, a constant pressure of 10¹⁵ K cm⁻³, and the boost of He II 303.786 Å.
• Peaks: the peaks agree with V10 to within 4% (6% in 211 Å).
• Where it differs (all differences come from CHIANTI's atomic data):
  • 94 Å is 4× lower at log T 6.35, because CHIANTI 11 moves the brightest Fe XIV line in the channel from 93.6 Å to 91.3 Å.
  • Fe VII, which CHIANTI 9.0.1 lacks, adds 18–49% to 171, 193 and 211 Å at log T 5.5.
  • Mg VII adds 22% to 94 Å at log T 5.8.
• Corrections: eve and time apply to both tables. chiantifix corrects SolarSoft's emissivity only, so it raises with "CHIANTI 11".
• Generator: tools/aia_temperature_response_chianti.py rebuilds the table, and rebuilds V10 from CHIANTI 9.0.1 to 0.1–3% rms.

Fixed: Filtergram.from_fits() and aia.open() with detector axes other than the defaults (#27). They raised KeyError, because Filtergram.empty() named the components of crpix and pc detector_x and detector_y whatever the axes were called.

What's Changed

• Fix Filtergram.from_fits() for detector axes other than the defaults by @roytsmart in #27
• Add an AIA temperature response computed from CHIANTI 11 by @roytsmart in #28

Full Changelog: v1.2.0...v1.3.0