With this update SKIRT issues a warning to the simulation log in the following cases:
The wavelength range of an SED or SED family does not fully cover the wavelength range configured for the source system (so that the specific luminosity will be zero outside the supported range).
The optical properties for a dust mix are not available for all wavelengths (potentially) used in the simulation. The values outside of the supported wavelength range are clamped to the nearest available value (except in radio range; see next section).
These extra log messages help the user discover/understand limitations that might otherwise go unnoticed. For example:
The spectrum for the Bruzual-Charlot SED family "ends" at 160 micron. While this is not a problem for regular applications, in some test simulations it might produce a strange kink in the output spectrum.
Recent dust mixes tabulate the optical properties up to 10 cm (e.g., all dust mixes based on DustEM tables, including DrainLi and THEMIS), but many older dust mixes have an upper limit of only 1 mm (e.g., the tables supplied by Draine and Min). This means that dust extinction should not be trusted at those longer wavelengths. Again, this is not a problem for most applications.
Dust properties at radio wavelengths
If the simulation wavelength range extends beyond 10 cm, any dust cross sections beyond this wavelength are now forced to zero. None of the currently provided dust mixes offers optical properties beyond 10 cm, and historically any values outside the supported range are clamped to the nearest available value. This would lead to substantially overestimated dust extinction in the radio wavelength range. This "hack" avoids this issue.
For reasons of upwards compatibility, with dust mixes that support a shorter wavelength range, the cross sections between the supported upper wavelength limit and 10 cm are still clamped to the nearest available value (i.e. the value at the supported upper wavelength limit).
Log messages
With this update SKIRT issues a warning to the simulation log in the following cases:
These extra log messages help the user discover/understand limitations that might otherwise go unnoticed. For example:
Dust properties at radio wavelengths
If the simulation wavelength range extends beyond 10 cm, any dust cross sections beyond this wavelength are now forced to zero. None of the currently provided dust mixes offers optical properties beyond 10 cm, and historically any values outside the supported range are clamped to the nearest available value. This would lead to substantially overestimated dust extinction in the radio wavelength range. This "hack" avoids this issue.
For reasons of upwards compatibility, with dust mixes that support a shorter wavelength range, the cross sections between the supported upper wavelength limit and 10 cm are still clamped to the nearest available value (i.e. the value at the supported upper wavelength limit).
Tests All functional tests still run.