New icloud=3 option is meant as an alternative to icloud=1, but only currently tested in combination with RRTMG shortwave and longwave radiation schemes. The existing icloud=1 option is nearly always a binary zero or one depending on prior existence of water or ice clouds, whereas this new addition follows Mocko and Cotton (1995) in applicaiton of the Sundqvist et al (1989) scheme while using a grid-scale dependent RH threshold, different over land versus ocean points, but no differencens (land v. ocean) at temperatures below -12C. The current thresholds were based on preliminary subjective evaluation with eastern Pacific Ocean tropical cyclone and stratocumulus cloud simulations (supporting HWRF) and are expected to change in future versions with additional analyses. In order to function properly with RRTMG radiation, there must be supposed amounts of liquid water content (LWC) and ice water content (IWC) considered under partial cloudiness conditions. For simplicity at this time, the amount follows adiabatic LWC/IWC with an entrainment factor and additional constraints, all of which are expected to require further tunings at a future time.
Only grid volumes not already containing a threshold amount of cloud water and/or ice are considered for partly cloudy LWC/IWC additions. In other words, a fractional cloudiness value of one remains in place for grid volumes already containing explicit cloud hydrometeors (produced by the grid-resolved microphysics scheme), whereas values greater than zero and less than one have temporary LWC/IWC temporarily added for purposes of cloud treatment in RRTMG and NOT considered as part of the LWC/IWC within any microphysics scheme. The LWC/IWC addition is neither advected nor diffused and is eliminated after consideration only by the RRTMG schemes. Furthermore, although it does not currently consider the parameterized convective cloud fraction or shallow cumulus LWC/IWC, future combinations are under consideration. Therefore, the new option is available for use together with any particular convective parameterization scheme (as well as none at all) as well as any microphysics scheme but should not be enabled if a particular microphysics scheme considers less than fully saturated conditions to lead to partial clouds with explicit LWC/IWC or else it would be potential
"double-counting." Additionally, if a convective or shallow-cumulus parameterization considers fractional cloudiness and LWC/IWC coupled with the RRTMG radiation scheme, then this new option should be avoided.
Contiributed by G. Thompson, NCAR-RAL, Spring 2015
