Precipitation without precipitation type

[j0nes2k]

Hello,

I am running the latest UEMS with a domain over Europe. I am postprocessing these files via GrADS and I am running into strange errors in my output files.

In my emsupp_cntrl.parm file, I am extracting the following variables:

Code: Select all

 (BUCKET_APCP_ON_SURFACE)                                   SCAL=(4.0) !  Period Accumulated Total Precipitation (Liquid Equivalent; kg/m^2) FNUM=(434)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
 (CPTYPE_ON_SURFACE)                                        SCAL=(1.0) !  Categorical Precipitation Types at Surface (Rain, Freezing Rain, Ice Pellets, Snow (Code table 4.222)) FNUM=(160)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)

...which leads to the following variables in my GrADS control file (generated via g2ctl.pl):

Code: Select all

APCPsfc   0,1,0   0,1,8,1 ** surface Total Precipitation [kg/m^2]
CFRZRsfc   0,1,0   0,1,34 ** surface Categorical Freezing Rain [-]
CICEPsfc   0,1,0   0,1,35 ** surface Categorical Ice Pellets [-]
CRAINsfc   0,1,0   0,1,33 ** surface Categorical Rain [-]
CSNOWsfc   0,1,0   0,1,36 ** surface Categorical Snow [-]

The simulation is running stable without any problems. However I sometimes get output where APCPsfc is greater than zero (meaning that there is precipitation available=, but none of the categorical precipitation type flags is set.

Did anyone of you experience the same problem? Am I missing something here, for example extracting the wrong parameters?

Thank you for your help and best regards!

Jonas

[meteoadriatic]

I think those depend on microphysics chosen but also on something else maybe. This problem arise all the time and I never tried to find solution because I use that:

Code: Select all

 (RAIN ON MDL SFCS    ) SCAL=( 6.0)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
 (SNOW ON MDL SFCS    ) SCAL=( 6.0)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)

instead. It will give you this:

Code: Select all

RWMRhy1   0,105,1   0,1,24 ** 1 hybrid level Rain Mixing Ratio [kg/kg]
SNMRhy1   0,105,1   0,1,25 ** 1 hybrid level Snow Mixing Ratio [kg/kg]

Those vars can be named slightly different depending on wgrib2 you use but you will be able to find them quickly. Also there is same thing for graupel (that one depends on MP!) and maybe something else that I don't use.

From there you can do many good things with grads.

[j0nes2k]

Hello,

thank you for your help! In my original emsupp_cntrl.parm file I cannot find the variables you mentioned, I only find the variables on pressure surface like for example:

Code: Select all

 (RWMR_ON_ISOBARIC_SFC)                                     SCAL=(-4.0) !  Rain Mixing Ratio on Pressure Surface (kg kg-1) FNUM=(183)
 L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
 (SNMR_ON_ISOBARIC_SFC)                                     SCAL=(-4.0) !  Snow Mixing Ratio on Pressure Surface (kg kg-1) FNUM=(184)
 L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
 (GRLE_ON_ISOBARIC_SFC)                                     SCAL=(-4.0) !  Graupel Mixing Ratio on Pressure Surfaces (kg kg-1) FNUM=(416)
 L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
 (HAILMR_ON_ISOBARIC_SFC)                                   SCAL=(-4.0) !  Hail Mixing Ratio on Pressure Surfaces (kg kg-1) FNUM=(1055)
 L=(00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)

Will using these make a difference? Also can you explain how I should interprete the output - does this give a percentage of available rain in relation to complete precipitation? E.g. a value of 0.3 indicates that 30% of the total precipitation for this time step is rain (and the other 70% are snow etc)?

Thank you for your help!

Jonas

[j0nes2k]

So I went on to try this solution. First I added the following parts in emsup_cntrl.parm (by the way, it seems the parameters were renamed between your version of UEMS and the latest one):

Code: Select all

# from: data/tables/post/grib2/emsupp_cntrl.HYBRID_ALL
 (RWMR_ON_HYBRID_LVL)                                       SCAL=(-4.0) !  Rain Water on Hybrid (Native Model) Surfaces (kg Kg-1) FNUM=(181)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
 (SNMR_ON_HYBRID_LVL)                                       SCAL=(-4.0) !  Snow Mixing Ratio on Hybrid (Native Model) Surfaces (kg kg-1) FNUM=(182)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
 (GRLE_ON_HYBRID_LVL)                                       SCAL=(-4.0) !  Graupel Mixing Ratio on Hybrid (Native Model) Surfaces (kg kg-1) FNUM=(415)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)
 (HAILMR_ON_HYBRID_LVL)                                     SCAL=(-4.0) !  Hail Mixing Ratio on Hybrid (Native Model) Surfaces (kg kg-1) FNUM=(1056)
 L=(10000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000)

...which resulted in the following variables getting written into my GrADS control file:

Code: Select all

GRLEhy1   0,105,1   0,1,32 ** 1 hybrid level Graupel [kg kg-1]
HAILMXRhy1   0,105,1   0,1,71 ** 1 hybrid level Hail Mixing Ratio [kg kg-1]
RWMRhy1   0,105,1   0,1,24 ** 1 hybrid level Rain Mixing Ratio [kg kg-1]
SNMRhy1   0,105,1   0,1,25 ** 1 hybrid level Snow Mixing Ratio [kg kg-1]

However it seems that also with this method the error persists:

Code: Select all

ga-> set lat 40
ga-> set lon 26
ga-> d crainsfc
Result value = 0
ga-> d csnowsfc
Result value = 0
ga-> d cicepsfc
Result value = 0
ga-> d cfrzrsfc
Result value = 0
ga-> d GRLEhy1
Result value = 0
ga-> d HAILMXRhy1
Result value = 0
ga-> d RWMRhy1
Result value = 0
ga-> d SNMRhy1
Result value = 0
ga-> d apcpsfc
Result value = 0.742766

As you can see there is precipitation, but there is no type set in any of the category types. Do you have any idea what happens here?

Best regards,

Jonas

[meteoadriatic]

Don't know. Which microphysics you use?

[j0nes2k]

Here is my model configuration:

Code: Select all

  ******************************************************************************

     Domain & Run Information

       Domain Type               :  Limited Area
       Primary Time Step         :  Adaptive
       Step to Output Time       :  Yes
       Grid dimensions (NX x NY) :  560 x 450
       Vertical Layers (NZ)      :  45
       Grid Spacing              :  7.00 km
       Top of Model Atmosphere   :  50mb
       Parent Domain             :  NA


     Timing Information

       Start Date                :  2017 Jan 16 06:00 UTC
       End Date                  :  2017 Jan 19 12:00 UTC
       Simulation Length         :  78 Hours
       Boundry Update Freq       :  03 Hours


     Data Output Information

       File Output Freq          :  01 Hour
       Output File Format        :  netCDF
       Adjust Output Times       :  Yes


     Model Physics

       Cumulus Scheme            :  None
       Shallow Cumulus Scheme    :  Off
       Microphysics Scheme       :  Lin et al. 5-Class SM
       PBL Scheme                :  Yonsei University
       YSU Wind Correction       :  No Wind Correction
       Gravitational Settling    :  No Gravitational Settling
       Land Surface Scheme       :  Noah 4-Layer LSM
       Snow Cover Physics        :  New School
       Number Land Categories    :  24 (USGS)
       Number Soil Categories    :  16
       Number Soil Layers        :  4
       Surface Layer Physics     :  Monin-Obukhov
       Sfc Heat & Moisture Flux  :  Model Computed u* and Sfc Fluxes
       Long Wave Radiation       :  RRTM
       Short Wave Radiation      :  Dudhia
       SW Update Between Calls   :  Update SW Radiation
       Short Wave Scattering     :  1 x10-5 m2/kg
       Cloud Effects             :  Fractional Cloud Effects
       Slope Radiation Effects   :  No Slope Effects
       Topography Shading        :  No Shading Effects


     ARW Core Model Dynamics

       Dynamics                  :  Non-Hydrostatic
       Time-Integration Scheme   :  Runge-Kutta 3rd Order
       Diffusion Sheme           :  Simple Diffusion
       6th-order Diffusion       :  No 6th-Order Diffusion
       Eddy Coefficient Scheme   :  2D 1st Order Closure
       Damping Option            :  No Damping
       W Damping                 :  W Damping On
       Horiz Momentum Advection  :  5th Order
       Horiz Scalar Advection    :  5th Order
       Vert Momentum Advection   :  3rd Order
       Vert Scalar Advection     :  3rd Order
       Sound Time Step Ratio     :  4 to 1 (Large TS)
       Moisture Advection Option :  Positive-Definite
       Scalar Advection Option   :  Positive-Definite
       Momentum Advection Option :  Positive-Definite
       TKE Advection Option      :  Positive-Definite

[meteoadriatic]

Try for test WSM6 (= option 6 for MP_PHYSICS). If result is the same then I don't know what is going on really.

[surgeon]

There is one option in namelist.input, that can block precc type. I also had this problem once.
Unfortunately, I do not remember which option is responsible for that.

[j0nes2k]

Hello,

I asked the same question to Robert, his response was: the categorical precipitation type is computed in relation to the instantanious precipitation. This may differ from the accumulated period precipitation. E.g. there was precipitation during the accumulation period, but at the exact timestep there is no precipitation any more. So this is a systematic behaviour that can be explained.

Best regards,

Jonas

[meteoadriatic]

With all due respect but this explanation does not have sense.

I can understand that there may be a lot of situations where there is accumulated precipitation but at time of output there is none at various grid points and because of that categorical precipitations will be = 0 at those points. But here we have situation that we do not have categorical precipitation over whole domain and over all output times, if I understood this correctly.