Calculate a large spectrum by part

In may be faster to calculate a full-range spectrum on several partial spectral ranges, and combine them.

Uses the MergeSlabs() function for that.

Starting from 0.11, RADIS introduces a sparse waverange implementation that should make it possible to directly compute a full range spectrum. See the HITRAN full-range example

OH-hitemp-radisdb-1000K-#5120 new quantities added: ['emisscoeff']
OH-hitemp-radisdb-1000K-#5376 new quantities added: ['emisscoeff']
OH-hitemp-radisdb-1000K-#5632 new quantities added: ['emisscoeff']
Spectrum Name:  OH-hitemp-radisdb-1000K-#5120//OH-hitemp-radisdb-1000K-#5376//OH-hitemp-radisdb-1000K-#5632
Spectral Quantities
----------------------------------------
   abscoeff     [cm-1]  (99,502 points)
   emisscoeff   [mW/cm3/sr/cm-1]        (99,502 points)
   radiance_noslit      [mW/cm2/sr/cm-1]        (99,502 points)
   emissivity_noslit    (99,502 points)
   absorbance   (99,502 points)
   transmittance_noslit         (99,502 points)
Physical Conditions
----------------------------------------
   Tgas                 1000 K
   Trot                 1000 K
   Tvib                 1000 K
   isotope              N/A
   mole_fraction        0.1
   molecule             {'OH'}
   overpopulation       None
   path_length          1 cm
   pressure_mbar        10000.0 mbar
   rot_distribution     boltzmann
   self_absorption      True
   state                X
   thermal_equilibrium  True
   vib_distribution     boltzmann
   wavenum_max          N/A cm-1
   wavenum_min          N/A cm-1
Computation Parameters
----------------------------------------
   NwG                  N/A
   NwL                  4
   Tref                 296 K
   add_at_used          cython
   broadening_method    voigt
   cutoff               1e-27 cm-1/(#.cm-2)
   dbformat             hitemp-radisdb
   dbpath               /home/docs/.radisdb/hitemp/OH-13_HITEMP2020.hdf5
   default_output_unit  cm-1
   folding_thresh       1e-06
   include_neighbouring_lines  True
   memory_mapping_engine  auto
   neighbour_lines      0 cm-1
   optimization         simple
   parfuncfmt           hapi
   parsum_mode          full summation
   profiler             N/A
   pseudo_continuum_threshold  0
   radis_version        0.13.1
   sparse_ldm           True
   spectral_points      N/A
   truncation           50 cm-1
   waveunit             cm-1
   wstep                0.1 cm-1
   zero_padding         N/A
Config parameters
----------------------------------------
   DEFAULT_DOWNLOAD_PATH  ~/.radisdb
   GRIDPOINTS_PER_LINEWIDTH_ERROR_THRESHOLD  1
   GRIDPOINTS_PER_LINEWIDTH_WARN_THRESHOLD  3
   SPARSE_WAVERANGE     auto
Information
----------------------------------------
   calculation_time     0.22960855498968158 s
   chunksize            None
   db_use_cached        True
   dxG                  0.13753507880165727
   dxL                  0.20180288881201608
   export_lines         False
   export_populations   None
   export_rovib_fraction  True
   levelsfmt            None
   lines_calculated     5,672
   lines_cutoff         21,775
   lines_in_continuum   0
   load_energies        False
   lvl_use_cached       True
   parfuncpath          None
   total_lines          27447
   warning_broadening_threshold  0.01
   warning_linestrength_cutoff  0.01
   wavenum_max_calc     N/A cm-1
   wavenum_min_calc     N/A cm-1
----------------------------------------


(0.0, 1.0)

from radis import MergeSlabs, calc_spectrum

spectra = []
for (wmin, wmax) in [(50, 3000), (3000, 7000), (7000, 10000)]:

    spectra.append(
        calc_spectrum(
            wmin,
            wmax,
            Tgas=1000,
            pressure=10,  # bar
            molecule="OH",
            path_length=1,
            mole_fraction=0.1,
            wstep=0.1,
            databank="hitemp",
            verbose=False,
        )
    )

s = MergeSlabs(*spectra, resample="full", out="transparent")
print(s)
s.plot("transmittance_noslit", wunit="nm")

import matplotlib.pyplot as plt

plt.ylim(0, 1)