Figures comparing the Chabrier distribution coefficients used in McKee & Offner 2010 (and formerly the defaults) to those from Chabrier 2003.
The differences in likelihood can be substantial, though likely most of the net effects are fairly negligible. Nonetheless, to maximize backward compatibility / minimize unforeseen downstream user error, I think it's best we keep the chabrier2005 form around. It's just as valid, just slightly less 'canonical'.
Figures comparing the Chabrier distribution coefficients used in McKee & Offner 2010 (and formerly the defaults) to those from Chabrier 2003.
The differences in likelihood can be substantial, though likely most of the net effects are fairly negligible. Nonetheless, to maximize backward compatibility / minimize unforeseen downstream user error, I think it's best we keep the
chabrier2005form around. It's just as valid, just slightly less 'canonical'.