===================================================== Forecast Verification Testing for Newly-released Code ===================================================== Prior to each WRF code major release, simulations are run to verify that new and/or modified code produces reasonable and expected results. Forecasts include each of a selected combination of physics and dynamics options. Results are then used as follows: * Output from simulations with the new options is compared to observations and analyses (NCEP GFS surface and upper air), subjectively and objectively. Domain averaged BIAS and RMSE are calculated for conventional fields. * Output from simulations with the new options is compared to a suite of well-tested physics options. * A baseline test, using code that includes the new options, but may not be using them, is compared to the same baseline test using the previous version of WRF. Those baseline options are: * RRTMG long- and shortwave radiation * Tiedtke cumulus parameterization * YSU PBL scheme * Noah surface scheme * WSM5 microphysics | | Below are the new or modified options tested for each release. | | | V4.5 Release ============ The test configuration is as follows: * **Test dates** : 1-28 February, 2017 and 1-28 May, 2017 * **Domain configuration** : CONUS domain with 15 km resolution * **Grid structure** : 415 x 325 horizontal grid points, 40 vertical levels, model top at 30 hPa * **Map projection** : Lambert-conformal * **Initial and boundary conditions** : NCEP GFS 0.25 degree analysis, 6-hourly intervals * **Temporal information** : each case is initialized at 00 UTC and runs for 48 hours * 26 cases are run for both the summer and winter dates | | | Verification ------------ * All results are compared to 0.25 degree GFS surface and upper air analyses. * Domain-averaged BIAS and RMSE are calcuted for conventional fields | | | | Tests ----- The following options were tested for this release: * The new package to compute two-moment prognostics for graupel/hail in Thompson scheme * The new k-epsilon-theta PBL scheme * Updated P3 scheme * Scale-aware Tiedtke scheme * Updated Mynn-EDMF scheme * Tropical Suite * Conus Suite * Conus Suite but run with NoahMP LSM scheme * Options used for climate simulation * Comparisn of NoahMP-V4.4 and NoahMP-V4.5 | | | | | V4.4 Release ============ The test configuration is as follows: * **Test dates** : 1-28 February, 2017 and 1-28 May, 2017 * **Domain configuration** : CONUS domain with 15 km resolution * **Grid structure** : 415 x 325 horizontal grid points, 40 vertical levels, model top at 30 hPa * **Map projection** : Lambert-conformal * **Initial and boundary conditions** : NCEP GFS 0.25 degree analysis, 6-hourly intervals * **Temporal information** : each case is initialized at 00 UTC and runs for 48 hours * 48 cases are run for both the summer and winter dates | | Verification ------------ * All results are compared to 0.25 degree GFS surface and upper air analyses. * Domain-averaged BIAS and RMSE are calcuted for conventional fields | | | | Tests ----- The following options were tested for this release: * updated Thompson microphysics and cloud fraction schemes * aerosol-aware Thompson scheme with new black carbon data * updated P3 1-ice category, 1-moment cloud water scheme * new rrtmg cloud overlap scheme * updated mynn sfclay scheme * WRF-solar EPS model * Run-time option of greenhouse gas | | | | | V4.3 Release ============ The test configuration is as follows: * **Test dates** : 1-28 February, 2017 and 1-28 May, 2017 * **Domain configuration** : CONUS domain with 15 km resolution * **Grid structure** : 415 x 325 horizontal grid points, 40 vertical levels, model top at 30 hPa * **Map projection** : Lambert-conformal * **Initial and boundary conditions** : NCEP GFS 0.25 degree analysis, 6-hourly intervals * **Temporal information** : each case is initialized at 00 UTC and runs for 48 hours * 48 cases are run for both the summer and winter dates | | | Verification ------------ * All results are compared to 0.25 degree GFS surface and upper air analyses. * Domain-averaged BIAS and RMSE are calcuted for conventional fields * Results from the CONUS suite tests are also verified against station observations and soundings archived in the GDAS historical dataset | | | | Tests ----- The following options were tested for this release: * The turbulence kinetic energy (TKE) and TKE dissipation rate based 1.5-order closure (EEPS) PBL parameterization * The updated version of P3 microphsyics scheme that includes a one-ice category, 3-moment ice option * The NTU (National Taiwan University) scheme that applies double moments for the liquid-phase and triple moments for the ice-phase hydrometeors together with the consideration for ice crystal shape and density variations. * The enhanced GWD parameterization developed by NOAA * The multilayer BEP and BEP+BEM urban canopy models (UCMs) coupled with the Yonsei University (YSU) planetary boundary layer parameterization * The new dynamic irrigation scheme implemented in NoahMP * The Implicit-Explicit Vertical Advection (IEVA) scheme | | | | | V4.2 Release ============ The test configuration is as follows: * **Test dates** : 1-28 January, 2017 and 1-31 May, 2017 * **Domain configuration** : CONUS domain with 15 km resolution * **Grid structure** : 415 x 325 horizontal grid points, 40 vertical levels * **Grid location** : 39N, -101W * **Map projection** : Lambert-conformal * **Initial and boundary conditions** : NCEP GFS 0.25 degree analysis, 6-hourly intervals * **Temporal information** : each case is initialized at 00 UTC and runs for 48 hours * 31 and 28 cases are run for the spring and winter dates, respectively .. note:: Two extra suites of 3km resolution runs are conducted for the period 1-31 May 2017 to test spectral bin microphysics and predicted particle property (P3) bulk microphysics. | | | Verification ------------ * All results are compared to 0.25 degree GFS surface and upper air analyses. * Domain-averaged BIAS and RMSE are calcuted for conventional fields * Results from the CONUS suite tests are also verified against station observations and soundings archived in the GDAS historical dataset * Precipitation from the 15 km and 3 km simulations is verified with StageIV observations. | | | | Tests ----- .. csv-table:: :escape: \ :header: Option, Radiation, Cumulus, Microphysics, Sfc Layer, Surface, PBL Conus 4.1.5 |br| Conus 4.2, RRTMG, Modified Tiedtke, Thompson, Janjic, Noah, MYJ Tropical 4.1.5 |br| Tropical 4.2, RRTMG, New Tiedtke, WSM6, Old MM5, Noah, YSU MSKF, RRTMG, MSKF, WSM5, Monin-Obukhov, Noah, YSU BMJ, RRTMG, BMJ, WSM5, Monin-Obukhov, Noah, YSU MYNN, RRTMG, MSKF, WSM5, MYNN, Noah, MYNN IRRI-3, RRTMG, MSKF, WSM5, Monin-Obukhov, Noah, YSU 3d TKE, RRTMG, Modified Tiedtke, WSM5, Monin-Obukhov, Noah, Off FARMS, RRTMG, Modified Tiedtke, Thompson, Janjic, Noah, YSU FSBM (3km), RRTMG, Off, FSBM, Monin-Obukhov, Noah, YSU P3 (3km), RRTMG, Off, P3, Monin-Obukhov, Noah, YSU | | | | | V4.1 Release ============ The test configuration is as follows: * **Test dates** : 1-28 June, 2015, and 1-28 January, 2016 * **Domain configuration** : CONUS domain with 15 km resolution * **Grid structure** : 401 x 261 horizontal grid points, 40 vertical levels, model top at 30 hPa * **Map projection** : Lambert-conformal * **Initial and boundary conditions** : NCEP GFS 0.25 degree analysis * **Temporal information** : each case is initialized at 00 UTC and runs for 48 hours * 48 cases are run for both the summer and winter dates. | | | Verification ------------ * All results are compared to 0.25 degree GFS surface and upper air analyses. * Domain-averaged BIAS and RMSE are calcuted for conventional fields * Results from the STD suite tests are also verified against station observations and soundings archived in the GDAS historical dataset. | | | | Tests ----- Records missing. This will be updated if they are found. | | | | | V4.0 Release ============ The test configuration is as follows: * **Test dates** : 1-28 June, 2015, and 1-28 January, 2016 * **Domain configuration** : CONUS domain with 15 km resolution * **Grid structure** : 401 x 261 horizontal grid points, 40 vertical levels, model top at 30 hPa * **Map projection** : Lambert-conformal * **Initial and boundary conditions** : NCEP GFS 0.25 degree analysis * **Temporal information** : each case is initialized at 00 UTC and runs for 48 hours * 48 cases are run for both the summer and winter dates. | | | Verification ------------ * All results are compared to 0.25 degree GFS surface and upper air analyses. * Domain-averaged BIAS and RMSE are calcuted for conventional fields | | | | Tests ----- .. csv-table:: :escape: \ :header: Option, HVC, Moist, Radiation, Cumulus, Microphysics, Sfc Layer, Surface, PBL Stdhyb-v3.9.1.1, on, on, RRTMG, Kfeta, WSM5, Monin-Obukhov, Noah, YSU Stdold-v3.9.1.1, off, on, RRTMG, Kfeta, WSM5, Monin-Obukhov, Noah, YSU Stdv40, on, on, RRTMG, Kfeta, WSM5, Monin-Obukhov, Noah, YSU Stdv40thm0, on, off, RRTMGK, Kfeta, WSM5, Monin-Obukhov, Noah, YSU Kiaps, on, on, RRTMGK, Ksas, WSM6, Monin-Obukhov, Noah, YSU Aercu, on, on, RRTMG, MSKF, Morrison TM Aero, Monin-Obukhov, Noah, YSU Rap, on, on, RRTMG, Grell-Freitas, Thompson Aero, MYNN, Ruc, MYNN Rap2, on, on, RRTMG, Grell-Freitas, Thompson Aero, MYNN, Noah, MYNN tropical, on, on, RRTMG, New Tiedtke, WSM6, Old MM5, Noah, YSU CONUS-v3.9.1.1, off, off, RRTMG, Modified TKE, Thompson, Monin-Obukhov, Noah, MYJ CONUS-v4.0, on, on, RRTMG, Modified TKE, Thompson, MYJ, Noah, MYJ | | | | |