The spectral emissivity file (usually named with a .ems
file extension)
is used by various DIRSIG optical properties, but is primarily associated
with the Classic Emissivity model that was first introduced in DIRSIG3.
The file can contain one or more emissivity curves, where multiple curve
files can be employed by the texture mechanism of the Classic Emissivity
to introduce spectral variability within a given material class.
Format Details
The first line of the file is the number of spectral curves in the file.
The next 91 lines are spectrally constant, angular weighting values for the emissivity for angles ranging from 0 to 90 degrees (inclusively, resulting in 91 values).
![]() |
The angular weighting values were introduced when early versions of
DIRSIG focused on broadband LWIR simulations. These weighting values
cannot be assumed to hold across all wavelengths. Therefore, most
emissivity files will have these values filled with 1.0 .
|
The curve data following the curve count and angular weighting header varies depending on the age of the file.
Old Format
The "old" format files had each curve sampled on fixed 100 wavenumber increments from 39850 to 350 wavenumbers, resulting in exactly 396 spectral points. The spectral samples are in decreasing order. The curves are separated by a line containing the a single value denoting the curve number (start with curve #0).
The following example is for a file containing 2 curves.
2 1.0 1.0 [lines deleted for documentation purposes] 1.0 1.0 0 39850 0.6865 39750 0.6654 39650 0.7123 [lines deleted for documentation purposes] 550 0.9098 450 0.9102 350 0.9187 1 39850 0.7465 39750 0.6954 39650 0.7524 [lines deleted for documentation purposes] 550 0.9198 450 0.9012 350 0.9076
New Format
The "new" format was added during the DIRSIG3 era to allow the user
to bring in curves with spectral resolution better than 100 wavenumbers.
The primary indicator that a file is a "new" format file is that the
curve delimiter is the string CURVE_BEGIN
. The units of the spectral
samples are in microns and the samples are in increasing order.
The following example is for a file containing 300 curves.
300 1.0 1.0 [lines deleted for documentation purposes] 1.0 1.0 CURVE_BEGIN 0.40000 0.992053 0.40084 0.991947 0.40169 0.991833 [lines deleted for documentation purposes] 14.1335 1.000000 14.2092 1.000000 14.2857 1.000000 CURVE_BEGIN 0.40000 0.985854 0.40084 0.985616 0.40169 0.985213 [lines deleted for documentation purposes] 14.1335 1.000000 14.2092 1.000000 14.2857 1.000000
![]() |
The spectral samples in a "new format" file do not need to be on regular intervals. |
![]() |
Each curve in a "new format" file can be sampled at different spectral locations. |