SDVoigt Absorption Coefficient Repeatable Errors

[lidardoug]

Hello! I think I've found two reproducible issues with absorptionCoefficient_SDVoigt. These might be user error, but I want to put them in front of the community just in case. Any guidance would be greatly appreciated.

In this first example, an extremely low pressure value (1e-5 atm) throws an IndexError whereas a slightly higher pressure doesn't (1e-4 atm). Newer versions of Numpy throw errors on mismatches of shape/size when binary-indexing arrays - which appears to be what's happening here. Older versions didn't care. Don't know if that could be part of the issue? Also, if you use absorptionCoefficient_Voigt insetad of absorptionCoefficient_SDVoigt, it appears to work just fine.

from hapi import *
db_begin('data')

fetch('CO2', 2, 1, 6350, 6375, ParameterGroups=[
      '160-char', 'Voigt', 'SDVoigt'])

nu, coef = absorptionCoefficient_SDVoigt(
    SourceTables='CO2',
    Environment={'T':117,'p':1e-05},# THIS PRESSURE VALUE THROWS THE ERROR BELOW
    #Environment={'T': 117, 'p': 1e-04},  # THIS PRESSURE VALUE WORKS
    WavenumberRange=[6350, 6375],
    WavenumberStep=0.001,
    HITRAN_units=True)

#   File "C:\ProgramData\Anaconda3\lib\site-packages\hapi\hapi.py", line 33898, in pcqsdhc
#     WR1_PART4[index_CPF] = WR1

# IndexError: boolean index did not match indexed array along dimension 0; dimension is 371 but corresponding boolean dimension is 370

User elizamanelli also posted an issue with this same pcqsdhc (partially-Correlated quadratic-Speed-Dependent Hard-Collision Subroutine) ... I don't have the expertise to tell if it's related.

In this second example, the absorption coefficients returned are all zeros. Similarly to the previous example, if you use absorptionCoefficient_Voigt insetad of absorptionCoefficient_SDVoigt, it works just fine.

from hapi import *
db_begin('data')

wnlowbound = 6045
wnhighbound = 6070
fetch('CH4',6,1,wnlowbound,wnhighbound,ParameterGroups=['160-char','Voigt','SDVoigt'])

nu, coef = absorptionCoefficient_SDVoigt( #this works fine if I just do Voigt instead of SDVoigt
    SourceTables='CH4',
    #Environment={'T':Tnow,'p':Pnow},
    WavenumberRange=[wnlowbound, wnhighbound],
    WavenumberStep=0.001,
    HITRAN_units=True)

# coef
# Out[2]: array([0., 0., 0., ..., 0., 0., 0.])
#sum(coef>0)
#Out[3]: 0

Again, any guidance would be greatly appreciated.

~Doug