Cooling rates are defined as d(Fu-Fd)/dz*1/(Cp*ρ) where
Fu and Fd are
the upward and downward spectral radiance fluxes, z is altitude,
Cp is the molar
heat capacity (default value CPKMOL_DEF in file
(phyadj_dat.f90 in RFM v5)
adjustable using *PHY
section) and ρ
molar density of air molecules.
Cooling rates can be averaged over output intervals using the
but the user has to multiply by the output
grid wavenumber interval in order to get K/day for that spectral interval.
The actual cooling rate calculation involves a quadratic fit to three
adjacent profile levels. If the levels are equally spaced, this reduces to a
simple difference in net fluxes (Fu-Fd) calculated at levels
above and below the output level.