Continuum And Feature Extraction tool (CAFE) model fits to galaxy II Zw 096

Description of '*_cube_parcube.fits' files: CAFE best-fit parameter cubes for the same aperture grids as the spectral cubes. Each file contains 11 extensions (HDUs 0-10): HDU 0 PRIMARY empty HDU 1 VALUE best-fit parameter value HDU 2 STDERR 1-sigma uncertainty on VALUE (NaN if fixed) HDU 3 VARY 1.0 if parameter was free in the fit, 0.0 if fixed HDU 4 MIN lower bound used in the fit HDU 5 MAX upper bound used in the fit HDU 6 EXPR global table of lmfit constraint expressions HDU 7 PARNAME global table of parameter names and indices HDU 8 APERTURE nx x ny table of aperture labels (same as spectral cube) HDU 9 COORDINATE nx x ny tables of aperture centre RA and Dec [deg] HDU 10 ROTATION aperture rotation angle [rad] Spatial grid indexing (Python / astropy, 0-based): - Aperture label i_j is stored in APERTURE[i, j] (same as spectral cube). - Parameter values are VALUE[p, i, j] (also STDERR, VARY, MIN, MAX). (first axis = parameter index p; i = step_indx; j = step_indy). - Parameter names are in PARNAME['PARNAME'][p]; global index in PARNAME['INDEX'][p]. - Constraint expressions (lmfit expr strings, or 'None') are in EXPR['expr'][p], with EXPR['idx'][p] = p. Parameter content: Parameters comprise continuum components (e.g. CIR_FLX, CIR_TMP, CLD_FLX, CLD_TMP), opacity terms (e.g. CO2_TAU, CORV_TAU), emission lines, and individual PAH/dust feature drude parameters (Peak, Wave, Gamma). The number of parameters differs between grids (727 for 3x3, 433 for 4x3, 265 for 10x6) because different CAFE input configurations were used per CRETA stage. Below is the logical content of each extension. ------------------------------------------------------------------------- Extension 0 (PRIMARY) ------------------------------------------------------------------------- Empty primary header. ------------------------------------------------------------------------- Extensions 1-5: VALUE, STDERR, VARY, MIN, MAX ------------------------------------------------------------------------- Columns Format Units Label Explanations ------------------------------------------------------------------------- cube I*4 mixed parameter 3D image array NumPy shape (Npar, nx, ny): ------------------------------------------------------------------------- Extension 6: EXPR ------------------------------------------------------------------------- idx I --- Index Parameter index (0 .. Npar-1) expr A24 --- Expression lmfit constraint expression linking parameters, or 'None' ------------------------------------------------------------------------- Extension 7: PARNAME ------------------------------------------------------------------------- INDEX J --- Index Parameter index (0 .. Npar-1) PARNAME nA --- Name CAFE parameter name ------------------------------------------------------------------------- Extension 8: APERTURE ------------------------------------------------------------------------- APERTURE nA --- Label nx x ny character array of names APERTURE[i,j] = "i_j" ------------------------------------------------------------------------- Extension 9: COORDINATE ------------------------------------------------------------------------- RA nD deg RA ICRS right ascension, nx x ny grid DEC nD deg Dec ICRS declination, nx x ny grid ------------------------------------------------------------------------- Extension 10: ROTATION ------------------------------------------------------------------------- rotation D rad Rotation Aperture rotation ------------------------------------------------------------------------- Example ------------------------------------------------------------------------- from astropy.io import fits with fits.open("gridExtraction_..._parcube.fits") as h: i, j = 1, 2 pnames = [s.strip() for s in h["PARNAME"].data["PARNAME"]] p = pnames.index("CIR_FLX") value = h["VALUE"].data[p, i, j] stderr = h["STDERR"].data[p, i, j] vary = h["VARY"].data[p, i, j] name = h["APERTURE"].data["APERTURE"][i, j] To rebuild a CAFE model spectrum at spaxel (i, j), use the CAFE software parcube2parobj() helper (CAFE/cafe_helper.py) with the matching spectral cube wavelength vector.