Modifying and extending input data for refractive indices
For all components except dust, the OPAC data base (Hess et al., 1998; Koepke et al., 1997) is based on Shettle and Fenn (1979). The corresponding refractive index tables are presented by Fenn et al. (1985), and extend to 40 microns. Our data for sulfate, black carbon and sea salt are directly taken from the OPAC data base. Our data from organic aerosols (OA) are based on the same source, but given at different wavelengths derived from ECHAM and extending only up to 15 microns. There is no reason to restict the wavelengths. It is recommended to include the full table (18-9b) from Fenn et al. in the refractive index input file, using the "water soluble" component for POM.
Our data for mineral dust are based on Haapanala et al. (JGR, 2012), extended with the OPAC values for wavelengths >= 4 micron. In particular, the imaginary part in the visible part of the spectrum is strongly reduced compared to OPAC. The same data are used in ECHAM-HAM v2 (Zhang et al., ACP, 2012; Raisanen et al., JGR, 2012). Other studies seem to support higher absorption in the visible (e.g. Otto et al., Tellus, 2009; Colarco et al., JGR, 2013). The measurements presented by Otto et al. indicate that the imaginary part at 550 nm is higher than ~0.002, whereas it is 0.0011 in the current input file.
- update and extend refractive index used for mineral dust, e.g. using the average data presented in Fig. 6 of Otto et al.
- for OA, include full OPAC table from Fenn et al.
- include scaling of the refractive index for BC to the value proposed by Bond and Bergstrom (2005) in the input file. Currently, this is done in the code.
Once this has been done, TM5 can also be used to calculate optical properties in the infrared part of the spectrum, for use in the longwave radiation scheme of IFS. Currently, these are calculated in IFS, based on mass concentrations from TM5.
Note that Fig. 1 in Stier et al. (2007) is wrong: the wavenumber order seems to be reversed.