

CIRC:
Continual
Intercomparison of Radiation Codes
The
output
(reference results) consists of the following:
 Pressure at atmospheric levels
 Upward broadband LW flux at atmospheric levels
from LBLRTM
 Downward broadband LW flux at atmospheric
levels from LBLRTM
 Broadband LW heating (cooling) rate of
atmospheric layers from
LBLRTM
 Central wavenumbers of 1 cm1 bins (boxcar) for
which LW spectral
output from LBLRTM is provided
 Downward spectral LW flux at surface within 1
cm1 bins (boxcar)
from LBLRTM
 Upward spectral LW flux at TOA within 1 cm1
bins (boxcar) from
LBLRTM
 Downward broadband total SW flux at surface
from CHARTS
 Downward broadband diffuse SW flux at surface
from CHARTS
 Upward broadband SW flux at TOA from CHARTS
 Central wavenumbers of SW spectral output from
CHARTS
 Downward total spectral (1cm1) SW flux at
surface from CHARTS
calculated using deltaM scaling
 Downward true
direct
(no contributions from forward scattered radiation) horizontal spectral
(1cm1) SW flux at surface from CHARTS
 Upward spectral (1cm1) SW flux at TOA from
CHARTS
Tips
and
instructions on reading and using the output files
The output for each CIRC case consists of 3
files:
 LW_lblrtm_1cm1_caseX.txt:
Contains fluxes at the boundaries of the atmospheric column
(downwelling surface and upwelling TOA) at 1 cm1
resolution from LBLRTM.
 LW_lblrtm_bb_caseX.txt:
Contains broadband flux and heating rate profiles from LBLRTM
 SW_charts_1cm1_caseX.txt:
Contains fluxes at the boundaries of the atmospheric column
(downwelling surface and upwelling TOA) at 1 cm1
resolution from CHARTS
The SW broadband fluxes are given at the beginning of the CHARTS 1 cm1
output files. Note that the LW flux and heating rate profiles are
inverted (values start at TOA and proceed to surface) in contrast to
the input profiles where surface values are placed first.
Here is a sample FORTRAN code to read
the contents of these files. The output files themselves can be
downloaded from this page.
Description
of
reference observations and calculations
Observations
The longwave
and shortwave surface irradiance measurements utilized for
the CIRC SGP and NSA cases are 5minute averages of the irradiance
values
provided by ARM Best
Estimate Radiative Flux VAP. The
5minute averaging window is centered at the time provided for each
case. The observed TOA irradiance values are from the temporally
closest CERES measurement for Case 4 (NSA) and the
temporally closest GOES
broadband fluxes inferred from a
radiance
to flux and narrowband to broadband conversion algorithm,
for cases 3 (SGP) and 7 (PYE). For the
remaining SGP cases the weighted
average
of the two GOESinferred fluxes bracketing
the nominal BBHRP calculation time were used. The observed broadband
irradiances can be found in the SW
and LW summary tables.
Reference
calculations
The reference
radiative transfer calculations are performed with CHARTS v. 4.04 and
LBLRTM
v. 11.1 for the SW and LW part of the spectrum, respectively. The
continuum model used was MT_CKD_2.0
and the line parameter file was AER_V_2.0.
LBLRTM is run
using 3 streams (i.e, 3 angles) while CHARTS is run
using 16 streams with deltaM scaling and in our configuration provides
fluxes only at the
top and bottom boundary of the atmospheric column. The surface albedo
is resolved at 1 cm1 (the same as the one provided to the
participants) and is linearly interpolated to the wavenumber of the
calculation. The
cloud optical properties (extinction coefficient, singlescattering
albedo and asymmetry parameter) for the two cloudy cases are calculated
from the retrieved layer water contents and effective radii using Mie
theory and
an assumed gamma droplet size distribution. Aerosols are treated similarly to
BBHRP, i.e., the spectral aerosol optical thickness is obtained from
the Angstrom relationship while g
and SSA are spectrally
invariant; a HenyeyGreenstein phase function is assumed. Output is
averaged at 1 cm1
intervals. In LBLRTM the cloud absorption optical thickness is also
obtained from Mie calculations and cloud scattering is not accounted
for (if
your model accounts for LW scattering you may want to turn it off, if
possible, for a more consistent comparison), as is any aerosol effect.
The
surface emissivity is set to unity across the LW spectrum .
Questions?
For
questions or problems reading/interpreting the output
dataset please contact Lazaros Oreopoulos.


