Description of KF-CuP Scheme (V3.8)

1) Short Description:

The KF-CuP parameterization of sub-grid scale clouds modifies the Kain-Fritsch (KF, Kain and Fritsch, 1990; Kain, 2004) ad-hoc trigger function with one linked to boundary layer turbulence via probability density functions (PDFs) using the cumulus potential (CuP) scheme (Berg and Stull, 2005; Berg et al., 2013).  An additional modification is the computation of cumulus cloud fraction based on the time scale relevant for shallow cumuli.  The parameterization includes a number of additional free parameters, including the number and size of bins used to define the PDF, the minimum frequency of a bin within the PDF before that bin is considered for shallow clouds to form (designed to neglect bins in the tail of the distribution that represent a very small fraction of the total PDF), and the critical cumulative frequency of bins required to trigger deep convection. However, the parameterization was found to be relatively insensitive to each of the parameters (Berg et al., 2013).  An evaluation of KF-CuP and its improvements in simulated downwelling shortwave irradiance compared to the default KF scheme is shown in Berg et al. (2013).

As described by Berg et al. (2015), KF-CuP includes an optional treatment of vertical transport of trace gases and aerosols, aqueous chemistry, wet removal, and aerosol effects on cloud drop number when coupled with the MOSAIC treatment in WRF-Chem for aerosols.  Treating the aerosol effects in parameterized, sub-grid scale clouds is important when the grid spacing is relatively large (greater than several km) and does not explicitly resolve shallow or deep convection.  The user has the ability to include the effects of parameterized clouds associated with either shallow or deep clouds or both.  Berg et al. (2015) used this option to show that the effects of vertical transport and cloud chemistry are reasonable.  The so-called indirect feedback on radiation is not yet implemented with KF-CuP, and additional tests are necessary before its inclusion in the model.

2) namelist information and recommendations

To run KF-CuP in WRF (meteorology only / no chemistry):

cu_physics = 10
shallowcu_forced_ra = ‘false’
It is recommended to set shallowcu_forced_ra to ‘false’.  Setting it to true will override the cloud fraction calculations to a prescribed maximum cloud fraction which can be changed by the user for sensitivity testing purposes.
shcu_aersosols_opt = 0; Flag to control aerosols in shallow convection.  Setting it to 0 indicates that aerosols are not included.
numbins = 21; Number of perturbations for potential temperature and specific humidity, should be an odd number
thBinSize = 0.1; bin size of potential temperature perturbation increment (K)
rBinSize = 1.0e-4; bin size of mixing ratio perturbation increment (kg/kg)
minDeepFreq = 0.333; minimum frequency required before deep convection is allowed
minShallowFreq = 1.0e-2; minimum frequency required before shallow convection is allowed

The values for numbins, thBinSize, rBinSize, minDeepFreq, and minShallowFreq are recommended values; however, there is flexibility for the users to alter these values. Berg et al. (2013) list the range of values that were used to test the sensitivity of the scheme.
It is strongly recommended to run KF-CuP with cu_rad_feedback = ‘true’, otherwise simulations will not include the feedback of t:qhe parameterized clouds on radiation and consequently other meteorological quantities.
KF-CuP has only been tested using for the CAM radiation scheme (ra_sw_physics = 3, ra_lw_physics = 3), although in theory it should work with any radiation package.

To run KF-CuP coupled with chemistry in WRF-Chem:

Same namelist variables as above but now including:

shcu_aersosols_opt = 2; Flag to control aerosols in shallow convection.  Setting it to 2 indicates that aerosols are included.
chem_conv_tr =1
mp_physics = 2 or 10
progn = 1
chem_opt = 7, 8, 9, 10, 203, or 204.  The 203 option is recommended since it is the version with the most up-to-date treatment of secondary organic aerosols (SOA) in MOSAIC.  Other chemistry namelist options will need to be changed for the associated value of chem_opt.

Having cu_physics = 10, shcu_aerosols_opt = 0, and chem_conv_tr = 1should not be used.  The code will run but the simulation may produce unexpected results.

References:

Berg, L.K, M. Shrivastava, R.C. Easter, J.D. Fast, E.G. Chapman, and Y. Liu, 2015: A new WRF-Chem treatment for studying regional scale impacts of cloud-aerosol interactions in parameterized cumuli. Geophys. Model Devel., 8, 409-429. DOI:10.5194/gmd-8-409-2015.
Berg, L.K., W.I. Gustafson, E.I. Kassianov, E. I., and L. Deng (2013), Evaluation of a modified scheme for shallow convection: Implementation of CuP and case studies, Mon. Wea. Rev., 141, 134-147, 10.1175/mwr-d-12-00136.1.
Berg, L.K., and R.B. Stull (2005), A simple parameterization coupling the convective daytime boundary layer and fair-weather cumuli, J. Atmos. Sci., 62, 1976-1988.
Kain, J.S., (2004), The Kain-Fritsch convective parameterization: An update, J. Appl. Meteor., 43, 170-181.
Kain, J.S., and J.M. Fritsch (1990), A one-dimensional entraining detraining plume model and its application in convective parameterization J. Atmos. Sci., 47, 2784-2802.