Skip to content

WRF Version 4.4
Compare
Choose a tag to compare
@kkeene44 kkeene44 released this 26 Apr 20:42
· 216 commits to master since this release
v4.4
c11bb76

The WRF Model has been updated to Version 4.4 on April 26, 2022.

The WRF Pre-processing System (WPS) has been updated to Version 4.4.

Acknowledgements: We would like to thank Do Ngoc Khanh (Tokyo Institute of Technology), Ted Mansell (NOAA/NSSL), Joseph Olson (NOAA/GSL), Mike Iacono and John Henderson (AER), Dongmei Xu (Nanjing University of Information Science and Technology), Wen-Jou Chen,Ying-Jhen Chen, and Yi-Chuan Lo (Central Weather Bureau, Taiwan), Alexander Ukhov (KAUST), Marc Honnorat (EXWEXs), David Wong, (US EPA), Jaemo Yang (NREL), Patricia Balle (HPE), Tzu-Chin Tsai (National Taiwan University), Mathieu Le Corre (University of Toulouse, France, and University of Brest, France), Jordan Schnell (NOAA), Rebecca Schwantes (CIRES/NOAA CSL), and Prasanth Valayamkunnath, Alma Hodzic, Davide Del Vento, Masih Eghdami, Soyoung Ha, Jamie Bresch, Hugh Morrison, Craig Schwartz, Cenlin He, Rajesh Kumar, Ryan Cabell, Ju Hye Kim, Greg Thompson, Maria Frediani, Patrick Hawbecker, Timothy Juliano, and Pedro Jimenez (NCAR), and David Gill.

New in Version 4.4

Physics

  • The NoahMP code has been moved to an external repository and is linked to WRF repository via a submodule. This submodule is automatically pulled in to build the code during compiling, as long as the compilation is done with internet access. If compiling without internet access, or you are not sure if the computer you are using to compile supports git (git is used during the compiling process to access code in submodule), or in the rare case that one checks out another branch, other than ‘master,’ it may be necessary to issue the following commands while connected to the internet to ensure the submodule is available for compiling.
    git clone https://github.com/wrf-model/WRF.git
    git submodule update --init --recursive
    Or in one command:
    git clone --recurse-submodule https://github.com/wrf-model/WRF.git

After this, it is okay to go offline to compile. In the case you would like to remove code in the submodule, use ‘clean -aa’ instead of ‘clean -a’. Details

  • The shallow-water roughness scheme from Jiménez and Dudhia (2018) is included for offshore roughness adjustment in water depths less than 100 m and activated by namelist option shalwater_z0 = 1. The option works with a specified depth or real bathymetry input (namelist list option shalwater_depth) (Details). The bathymetry data is available from WPS/geogrid. Please acknowledge the following presentations and publications:
    [1] GEBCO Compilation Group (2021) GEBCO 2021 Grid, DOI: 10.5285/c6612cbe-50b3-0cff-e053-6c86abc09faf
    [2] Jiménez, P. A., and J. Dudhia (2018). On the need to modify the sea surface roughness formulation over shallow waters. J. App. Meteor. and Clim., 57(5), 1101-1110, DOI: 10.1175/JAMC-D-17-0137.1

  • WRF-Solar is expanded to have a stochastic ensemble prediction system (WRF-Solar EPS) tailored for solar energy applications (Yang et al., 2021, Kim et al., 2022). The stochastic perturbations can be introduced into variables of six parameterizations controlling cloud and radiation processes (Details). A more detailed description of the model is provided on the WRF-Solar EPS website.
    References:
    [1] Kim, J.-H., P.A. Jimenez, M. Sengupta, J. Yang, J. Dudhia, S. Alessandrini, and Y. Xie, 2022: The WRF-Solar Ensemble Prediction System to provide solar irradiance probabilistic forecasts. IEEE J. of Photovoltaics, 12, 141-142.
    [2] Yang, J., J.-H. Kim, P.A. Jimenez, M. Sengupta, J. Dudhia, Y. Xie, A. Golnas, and R. Giering, 2021. An efficient method to identify uncertainties of WRF-Solar variables in forecasting solar irradiance using a tangent linear sensitivity analysis. Solar Energy, 220, 509-522.

  • Greenhouse gas (GHG) concentrations from a number of RCPs and newer SSPs are now a default run-time option (previously they were a compile-time option). This serves two purposes: 1) Since data files provide compiled global climatological values for co2, n2o, ch4, cfc11, and cfc12, up to 2006 for RCPs and up to 2014 for SSPs, they are better estimates for historical and current runs, 2) If users have values of their own, they can be easily added to the data file. The user specifies “ghg_input=1” in the &physics namelist record for climatology, which is now the default, or “ghg_input=0” for constant values for backward compatibility. The default file used is CAMtr_volume_mixing_ratio.SSP245. A simple function for CO2 is now the default when choosing not to use the climo GHG files for RRTM - previously this function was only in RRTMG schemes. This option is only available for radiation schemes CAM, RRTMG, RRTMG-Fast, and RRTM. Details

  • The option is added to output 16 accumulated physics tendencies for potential temperature, water vapor mixing ratio, and u and v components of wind. This is turned on with the namelist option “acc_phy_tend=1” in &physics. Details

Other New Options

  • The capability to couple WRF and CMAQ (an air quality model developed at the US EPA) is added. As a coupled model, WRF-CMAQ can be run in a one-way fashion (i.e., meteorological information will be transferred to CMAQ directly to drive its calculation), or in a two-way fashion, (i.e., the aerosol information from CMAQ will be fed back to the RRTMG radiation physics in WRF). This is available beginning in WRFV4.4 and CMAQ v5.3.3.3. Refer to the EPA’s WRF-CMAQ Model webpage for instructions. Details

  • The ability to write compressed NetCDF4 files in parallel, via NetCDF V4.7.4 (and later) is added. Performance is slower than pnetcdf, but can be notably faster than regular NetCDF on parallel file systems. As expected, compression provides files significantly smaller than pnetcdf generates. Details

Improvements and Bug Fixes

Physics

Microphysics

  • Thompson Aerosol-aware microphysics: Code enhancements are introduced to the processing of aerosol information in real.exe; specifically, legacy code to support the old format aerosol data before 3.9 is removed and logic in the real program is improved. Details

  • Thompson Aerosol-aware microphysics: A black carbon aerosol category is added. To include black carbon in this microphysics as well as its radiative effect, an updated data file, QNWFA_QNIFA_QNBCA_SIGMA_MONTHLY.dat, is required and set wif_input_opt = 2 (option 1 works with water/ice friendly aerosol data only). Support is also added for ingesting real-time data that contains these aerosols (such as from GEOS-5) by setting use_rap_aero_icbc = .true. (Details). Time-varying surface aerosol emission from either climatology or real-time data is now handled using namelist option qna_update = 1 with input file name wrfqnainp_d0* (must set auxinput17_interval and io_form_auxinput17). In addition, biomass burning organic and black carbon aerosols can be added to the model by setting namelist wif_fire_emit = .true.. This option will work with data with and without black carbon, but it is particularly important during active wildfire hence when using a first-guess aerosol source that has information about biomass burning emissions (e.g., GEOS-5). When using this option, it is assumed this aerosol is evenly distributed in the PBL (controlled by namelist wif_fire_inj = 1). Details

  • Thompson microphysics: An update to this, plus the cloud-fraction scheme “icloud=3” to better match observations. Modifications include updates to RRTMG LW and SW, and RRTMG fast LW and SW. Details

  • P3 microphysics: Four options (50, 51, 52, and 53) of the P3 scheme are updated. The 3-moment lookup table file size is reduced by an order of magnitude (file names are changed for the lookup tables). Improved flexibility has been added by including the Kogan (2013, JAS) option for cloud autoconversion/accretion. A few minor bug fixes to harden the code have been added (Details).
    References:
    [1] Morrison, H., and J. A. Milbrandt, 2015: Parameterization of cloud microphysics based on the prediction of bulk particle properties. Part 1: Scheme description and idealized tests. J. Atmos. Sci., 72, 287-311.
    [2] Milbrandt, J. A., and H. Morrison, 2016: Parameterization of cloud microphysics based on the prediction of bulk particle properties. Part 3: Introduction to multiple free categories. J. Atmos. Sci., 73, 975-995.
    [3] Milbrandt, J. A., H. Morrison, D. T. Dawson II, and M. Paukert, 2021: A triple-moment representation of ice in the Predicted Particle Properties (P3) microphysics scheme. J. Atmos. Sci., 439-458.

  • NTU microphysics: Fixed uninitialized variable BHTMP. Details.

  • NSSL microphysics: Fixed a few uninitialized variables. Details

Radiation

  • A new cloud overlap option is available for RRTMG radiation schemes. The option is set by using the exponential-random namelist parameter “cldovrlp=5.” A decorrelation length option, “idcor,” is also added. It can be set as a fixed value at 2500 m (idcor=0), or a function of latitude (idcor=1). The decorrelation option also works for cloud overlapping option 4 (exponential). Details

  • KIAPS Longwave Radiation: In V4.2, an error was introduced when attempting to fix a compilation failure, but ended up introducing yet another compile-time failure. The solution is to revert the entire KIAPS longwave radiation scheme to the original pristine condition. Details

PBL

  • Shin-Hong PBL: A minor bug fix to avoid a rare divide by zero issue. Details

LSM

  • NoahMP LSM: Updated with the following new development: Details
  1. Added three runoff schemes (opt_run = 6,7,8).
  2. Added a new drainage scheme (opt_tdrn)
  3. Added new soil step control capability (namelist soiltstep)
  4. Added water and energy budget terms for output and diagnosis (namelist noahmp_output)
  5. Updated flood irrigation parameters
  6. Added constraints for snow thickness during compaction
  7. Updated snow viscosity coefficient in compaction process
  8. Added canopy heat storage in canopy temperature calculation
  9. Updated formulation and parameter values for TOPMODEL groundwater scheme (opt_run=1)
  10. Bug fixes for SWE cap value in initialization, NoahMP snow depth update, snow layer index update, floating issues in runoff scheme, and BATS albedo scheme.
  11. NoahMP Glacier: An issue is resolved that occurred when there was no sunlight (cosz<0), causing the NoahMP BATS snow albedo calculation using cosz to give a “divided by zero” error in glacier code. Details
  12. NoahMP: Corrected declaration for two variables, FV and USTAR, from ‘out’ to ‘inout’ ones. Details

  • A fix addresses a runtime segmentation fault when using “sf_surface_mosaic=1” with “use_wudapt_lcz=0.” Details

  • Noah LSM: Fixed memory error that may cause occasional model failure when WRF-Hydro is not used. Details

  • Noah LSM: Fixed two input variables of absorbed longwave radiation LWDN and net shortwave radiation SOLNET for non-urban fraction of the LSM call. Details

  • RUC LSM: The default method to diagnose 2 m temperature and mixing ratio is changed to the flux method in RUC LSM, the same as used in NCEP operational RAPv5/HRRRv4. Details

Surface Layer

  • Updates and bug fixes to the MYNN surface layer scheme to bring this scheme closer to the CCPP version (Details), described in Olson et al. (2021):
    Olson, Joseph B., Tanya Smirnova, Jaymes S. Kenyon, David D. Turner, John M. Brown, Weizhong Zheng, Benjamin Green (June 2021): A description of the MYNN surface layer scheme. NOAA Technical Memorandum. OAR GSL-67. pp. 33., DOI:10.25923/f6a8-bc75

Urban

  • A minor bug fix for urban wind directions when using detailed urban morphology maps. Details

  • An issue that caused an underestimation of the rain rate used in urban physics is resolved. Details

  • Fixed an uninitialized emissivity variable in Noah driver for urban option. Details

Orographic Gravity Wave Drag

  • The description for the 10 orographic gravity wave drag variables (gwd_opt=1 only) are updated. Details

Others

  • Occasionally when running real.exe on multiple processes with MPI, one or more processes crashed with error “Unrecognized physics suite.” A fix for the MPI synchronization bug related to (not used) split communicators in the real program provides a solution to this issue. For users that had no trouble with this problem, this fix will have no impact. Details

  • To address an issue with compiling WRF on AWS cloud computing platforms (and perhaps other OSs), with Intel, a work-around is introduced to include the unix command “true” as the last command in a string of registry-based commands. This allows WRF to build on AWS with the Intel compiler. For other users, this will have no impact. Details

  • The configure file is modified to address a compiling error when compiling the vortex-following “hires” option related to the landread.c file. Details

  • Beginning with V4.2.2, a bug was introduced that prevented users from configuring the model with the “INTEL (ifort/icc): HSW/BDW” option. This is now resolved. Details

  • The NMM dynamical core and some NMM-specific source code are removed from the WRF repository. Details

  • A new variable that accumulates GHI over time is added to the WRF-Solar diagnostics package. Cloud base and cloud height units (AGL) are clarified in the registry. Details

WRF-Fire

  • Added the ability to customize vertical levels for fire runs. Vertical nesting is still not supported. Details

  • A new module to parameterize firebrand spotting for WRF-Fire is added. This is a passive Lagrangian transport scheme to transport and burnout firebrands generated at the fire front. The scheme is activated when ifire==2 by setting the namelist option fs_firebrand_gen_lim to an integer greater than zero (default is 0, i.e., scheme is off). It runs with dmpar- and serially-compiled code and in the inner nest (grid_id==max_dom). It was designed and tested using a mesoscale to LES domain configuration. Details

WRFDA

  • WRFDA is now able to ingest NCEP’s satwnd.bufr (gdas.satwnd.tHHz.YYYYMMDD.burf) to assimilate more AMVs that are not included in NCEP’s prepbufr files. More flexible thinning/superobbing options (thin_conv_opt) and obs error specification (uv_error_opt) are implemented for using satwnd.bufr. See WRFDA/var/README.namelist for details. Details

  • A new capability to assimilate GPM-GMI radiance data is introduced.
    Reference:
    Shen, et al., 2021: Assimilation of GPM Microwave Imager Radiance data with the WRF Hybrid 3DEnVar System for the Prediction of Typhoon Chan-hom (2015), Atmospheric Research. 251, 105422. Details

  • WRFDA is newly-expanded for “chem_cv_options=108” for surface chemical data assimilation (PM2.5, PM10, CO, SO2, NO2, O3) using the RACM/MADE-VBS chemistry option in WRF-Chem. Details
    Reference:
    Ha, Soyoung, 2021: Implementation of aerosol data assimilation in WRFDA (V4.0.3) for WRF-Chem (V3.9.1) using the MADE/VBS scheme. Geoscientific Model Development. gmd-2021-74.

  • The WRFDA code is corrected to reduce memory usage for “ob_format=1” (i.e., bufr for conventional obs) and “thin_conv=.true.” Details

  • WRFDA is fixed and enhanced for applications related to offline varbc (use_background_error=.false.), 3DEns-FGAT (num_fgat_time>0 and ensdim_alpha>0 and je_factor=1.0), precondition_cg=.true. and orthonorm_gradient=.true. when ensdim_alpha>0, full-ensemble mode (ensdim_alpha>0 and je_factor=1.0). Details

  • WRFDA gfsref code is restructured to move QC steps in da_get_innov_vector_gpsref.inc to a separate subroutine in a new file, “da_qc_gfsref.inc.” New namelist variables are added to allow more flexibility for testing gpsref QC impact. Several issues with gpsref diagnostic output have been fixed and qcstat information for gpsref is now in new qcstat_gpsref_xx. Details

  • Modifications to address compilation errors due to aqueous variables in da_get_innov_vector_chem_sfc.inc. Details

  • Fixed an issue that occurred when assimilating with pm2.5 (e.g., chemicda_opt = 3 or 5), using chem_opt=108. pm10 is treated as (pm10 - pm2.5) residuals in both observations and priors. PM10 obs errors are also adjusted for such residuals. Details

  • Bug fixes for direct assimilation of radar reflectivity. The namelist parameter “qthres,” which is used for the hydrometeor base state of TL/AD for no-rain background, should not be used in the reflectivity forward operator. QC for “radar_rf_opt=2” should not be confused with that of retrieval radar DA. Details

  • An issue is resolved related to inconsistency between TAMDAR and Radiance VarBC code. Details

WRF-Chem

  • Subgrid boundary layer clouds from the MYNN PBL scheme are now allowed to be accounted for in photolysis, using “phot_opt=4” (TUV). This feature is activated by setting the &chem namelist option “phot_blcld=.true.” “icloud_bl=1” must also be set. Details

  • Minor bug fix for aerosol optics Mie extrapolation cases in chemistry. When aerosol refractive indices are outside the lookup table ranges, extrapolation in the parameterized Mie calculation will create unrealistic values for the extrapolated aerosol optical parameters. These unphysical fluctuations cause unrealistically high AOD values and the model crashes. The solution is to force the out-of-range aerosol refractive index to be the minimum or maximum boundary values of the lookup table. Details

  • Modifications address an issue in which HONO, TERP and CO2 were not automatically pulled into the RACM mechanisms, even though they were in the anthropogenic emissions file. Details

  • An issue is corrected related to inconsistent data units for gas species (chem_cv_options). Details

WRF-Hydro

  • WRF-Hydro has been updated to version v5.3. Details

If downloading the files from the section below, please choose either the v4.4.tar.gz file, or the v4.4.zip file. DO NOT choose those named "Source Code." They do not include the mandatory NoahMP submodule - needed for compiling WRF.

Assets 4
Loading