Normal solar irradiance at top of atmosphere (TOA). Either use the solar spectrum (Modtran 3.5) provided at http://www.ecd.bnl.gov/~halthore/intercomp/intercomparison.html, or normalize your integrated value to 1373.12 W/m^2 (i.e., that of Modtran 3.5).
Assume that the TOA is the uppermost LEVEL in the data files.
In order to make sun angle dependent plots, each experiment will be evaluated at 20 values of cosine of solar zenith angle. We assume that most, if not all, of you will have your code in a large outer-loop thereby making it easy to do many sun angles during a single execution.
Model atmospheres [experiment names appear bold in parentheses, eg. (CLEAR), (CLOUD_A), etc.]
Clear sky reference case
cosine of SZA = 0.05 to 1.0 (in increments of 0.05), surface albedo = 0.2 (spectrally independent; Lambertian), atmosphere = standard Tropical (McClatchey et al.) (up to 101.737 km)
gaseous attenuation by (especially the first 3):
- Rayleigh
- H2O vapour
- O3
- CO2 [360 ppm]
- O2
(CLEAR)
Total: 1 case, 20 results.
Homogeneous overcast clouds reference cases
cosine of SZA = 0.05 to 1.0 (in increments of 0.05), surface albedo = 0.2 (spectrally independent; Lambertian), atmosphere = standard Tropical (McClatchey et al.) (up to 101.737 km)
gaseous attenuation by (especially the first 3):
- Rayleigh
- H2O vapour
- O3
- CO2 [360 ppm]
- O2
only liquid clouds are considered:
Initially we suggested that the optical properties should follow Slingo's
(1989) parameterization using effective radius for pure, spherical
droplets of 10 micrometers. We feel now, however, that this could be too
restrictive. Therefore, we are suggesting that you use whatever
parameterization or exact solution you wish but try to stick to pure,
spherical droplets of effective radius 10 microns. If need be, use
Henyey-Greenstein scattering phase functions.
(CLOUD_A) overcast low cloud positioned between 3.5 km - 4 km (LAYER 57 of the provided standard tropical atmosphere);
- mixing ratio = 0.159 g/kg --> visible optical depth of ~10.
(CLOUD_B) overcast high cloud positioned between 10.5 km - 11 km (LAYER 43 of the provided standard tropical atmosphere);
- mixing ratio = 0.034 g/kg --> visible optical depth of ~1.
Total: 2 cases, 40 results.
NOTE: These mixing ratio values are NOT in the "standard tropical" file that is provided. It is probably more straightforward to have you insert these values manually into your own code (and set cloud fraction to 1).
CRM fields
cosine of SZA = 0.05 to 1.0 (in increments of 0.05)
surface albedo = 0.2 (spectrally independent; Lambertian)
clouds and below: CRM data
above clouds and ozone profile (up to ~100 km): standard tropical
gaseous attenuation by (especially the first 3):
- Rayleigh
- H2O vapour (as provided in files)
- O3 (McClatchey Tropical)
- CO2 [360 ppm]
- O2
liquid clouds: as described in previous section (i.e., pure spheres with effective radius of 10 microns everywhere);
ice clouds: Ice clouds will be treated as though they are liquid to avoid obvious complications. We know this will be viewed by some as sacrilegious but the optical properties of ice crystals are so varied and details on what parameterizations are used by who are very sketchy. More details about the effects of treating ice crystals as water droplets can be seen at the bottom of this page.
Six diverse atmospheres generated by 3D CRMs have been selected based on recommendations from the GEWEX-GCSS activity (click to pop up a window containing a description and images of the cloud fields), (liquid water paths plots are low by a factor of 1000 -- conversion error):
The plots below demonstrate our rationale for neglecting rain, snow,
and graupel (i.e., precipitation) and including just liquid droplets
and ice crystals. These are the results from a broadband Monte Carlo
algorithm acting on 120 snapshots from a cloud resolving model's
simulation of a tropical cloud system (every 5 model minutes from
0700 to 1700; July 15 at the equator). All results presented are domain
averages. Fig. 1 shows how profiles of cloud fraction evolve over
time for just the "cloud" and the "cloud" plus "precip". Clearly, the
precip field is extensive. Fig. 2, however, shows how TOA albedo,
surface absorptance, and atmospheric absorptance evolve. Clearly, the
precip field has minor impact. Fig. 3 shows that the precip field
has a non-negligible influence on mean heating rate but considering
its areal extent, it is still rather minor. Thus, since it is unclear
how we should prescribe the optical properties of precip so as to
satisfy everyone, we have decided to neglect precip altogether.
References
Barker, H. W. and Q. Fu, 1998: Modelling domain-averaged solar fluxes for
an evolving tropical cloud system. Submitted to: Atmospheric and
Oceanic Optics (G. Titov Memorial).
Barker, H. W., G. L. Stephens, and Q. Fu, 1998: The sensitivity of
domain-averaged solar fluxes to assumptions about cloud geometry. In
press, Q. J. Royal Met. Soc.
Fu, Q., S. K. Krueger, and K.-N. Liou, 1995: Interactions of radiation
and convection in simulated tropical cloud clusters. J. Atmos. Sci.,
52, 1310-1328.
The treatment of ice crystals as water droplets
For an example showing the impact of ice crystals treated as though they are liquid droplets, click on the button below. For the most part, differences are rather small. Thus, to simplify matters, when considering liquid + ice and use effective radius of 10 microns everywhere... We feel that this does not compromise the integrity of the study for the prime objective of this phase of ICRCCM-III is representation by 1D models of unresolved clouds, not microphysical details.
