Ristic microphysics scheme at WRF model

Ristic microphysics scheme contributed by Weather2 has been accepted and tested in WRF version 4.4.1. The code passed all stages of testing. The code is in the public domain and can be downloaded from the official Weather2 Github repository:

Release WRF Version 4.4.1 (Ristic micro+conv) · Weather2/WRF (github.com)


The problem of description condensation-cloud-precipitation process are discussed for many years and has been the subject of various researches and scientific papers. Models that have radiation effect included, and thus need information on fractional cloud cover, describe condensation with diagnostic formulations that often give little regard to consistency with model-produced condensation fields. Sundqvist, et al. (1988) had proposed parameterization scheme for convective and stratiform condensation that is using cloud water and cloud fraction as prognostic variables, which gave good results when it comes to cloud water and precipitation forecasts. Formulation of fractional cloud cover depended only on the relative humidity, due to that, amount of low fraction cloud cover were unpredicted. Also, the amount of model cloud water had magnitude that possibly was unrealistically large. Because of this inconsistency, formula for fractional cloud cover did not produce good results when used in radiation scheme. This was the reason why the diagnostic formulation started to be used again.

This new scheme is based on using three prognostic equations fractional cloud cover, cloud mixing ratio and snow per cloud fraction. New fractional cloud cover formula, in addition to relative humidity, has cloud mixing ratio included. Adding new prognostic variable into the equation gives much more realistic description of cloud cover. Clouds predicted like this can be used in the model radiation calculations.

A complete description is now found in Ristic I., Kordic I., 2018: Cloud parameterization and cloud prediction scheme in the Eta numerical weather model. NWCM – Serbian Academy of Sciences and Arts which can be found at: https://www.sanu.ac.rs/wp-content/uploads/2018/10/11_Cloud%20parameterization%20and%20cloud%20prediction%20scheme%20in%20Eta%20numerical%20weather%20model.pdf

also at ResearchGate:




Please use this address for direct questions/comments about this page:

Mr. Ivan Ristic

Description of changes

This is list of modified files:














User’s guide

Users are encouraged to use WRF version 4.4.2 or upcoming versions.

In order to use this new mycrophysics after downloading and compiling the code following changes in namelist.input are required:

mp_physics = 150,

Fallowing changes are recomended:

cu_physics = 1,

ra_lw_physics                 = 99, 1, 1,

ra_sw_physics                 = 99, 1, 1,

Example of the namelist.input for one domain for the physics part:


!——————- MICROPHYSICS ———————-

mp_physics                    = 150,

!——————- RADIATION ————————-

ra_lw_physics                 = 99,

ra_sw_physics                 = 99,

radt                          = 60,

icloud                        = 0,

co2tf                         = 1,

!——————- SURFACE LAYER ———————

sf_sfclay_physics             = 1,

iz0tlnd                       = 0,

isftcflx                      = 0,

!——————- LSM ——————————-

sf_surface_physics            = 4,

num_soil_layers               = 4,

num_land_cat                  = 21,

!——————- PBL ——————————-

bl_pbl_physics                = 1,

bldt                          = 0,

!——————- CUMULUS —————————

cu_physics                    = 1,

cudt                          = 0,

Test results

Bellow are presented two real test cases in which the applied microphysics was tested.

First test situation for 17th November, 2011.

Because fog is very difficult parameter for forecasting we presented different parameters that are closely related to fog such as: predicted cloud cover and surface visibility, temperature on 2m and 850mb. Comparing model output with actual measurements from SYNOP reports and soundings we came up to conclusion that the WRF model with the new cloud prediction scheme showed good results when forecasting fog in northern Italy, Serbia and Croatia.

Figure 1. Satellite cloud picture at 18.11.2011 at 11 UTC


Figure 2. Predicted cloud cover from WRF model at 18.11.2017 at 12 am UTC
Figure 3. Predicted surface visibility (m) from WRF model at 18.11.2017 at 9 am UTC

Second test situation from 24th June, 2015. referred to precipitation forecast.

Predicted cloud cover showed very similar representation of cloud cover relative to satellite cloud picture for selected date and time. Subjective comparison shows that shape of cloudiness over Serbia is very similar to satellite image for the same area. Improvement in precipitation forecasts are also noticed, comparing predicted accumulated precipitation with measurements from GSOD data.

The quality of the precipitation forecast is important test of any cloud parameterization scheme, since clouds and precipitation are directly linked parameters.

Figure 4. Predicted 6h accumulated precipitation (mm) from WRF model (mm/6h) at 24.6.2015. at 12 UTC

Figure 5. 24h accumulated precipitation (mm) from GSOD data at 25.6.2015 at 00 UTC