Nesting Eta Model Ptop 0 Eta Coordinate And

Nesting

Eta Model

Ptop = 0 Eta Coordinate And Step Mountains ground MSL =1

Horizontal resolution of 12 km 12 -km terrain

Drawbacks of the Eta Coordinate • The failure to generate downslope wind storms in regions of complex terrain • Weak boundary layer winds over elevated terrain when compared to observations • The displacement of precipitation maxima too far toward the bottom of steeply sloping terrain as opposed to the observed location near the top half of the terrain slope • The reduction in the number of vertical layers used to define the model atmosphere above elevated topography particularly within the boundary layer

WRF Model Family A Tale of Two Dynamical Cores

Why WRF? • An attempt to create a national mesoscale prediction system to be used by both operational and research communities. • A new, state-of-the-art model that has good conservation characteristics (e. g. , conservation of mass) and good numerics (so not too much numerical diffusion) • A model that could parallelize well on many processors and easy to modify. • Plug-compatible physics to foster improvements in model physics. • Designed for grid spacings of 1 -10 km

Two WRF Cores • ARW (Advanced Research WRF) • developed at NCAR • Non-hydrostatic Numerical Model (NMM) Core developed at NCEP • Both work under the WRF IO Infrastructure NMM ARW

The NCAR ARW Core Model: (See: www. wrf-model. org) § Terrain following vertical coordinate § two-way nesting, any ratio § Conserves mass, entropy and scalars using up to 6 th order spatial differencing equ for fluxes. Very good numerics, less implicit smoothing in numerics. § NCAR physics package (converted from MM 5 and Eta), NOAH unified land-surface model, NCEP physics adapted too

The NCEP Nonhydrostatic Mesoscale Model: NMM (Janjic et al. 2001) § § § Hybrid sigma pressure vertical coord. 3: 1 nesting ratio Conserves kinetic energy, enstrophy and momentum using 2 nd order differencing equation Modified Eta physics, Noah unified land-surface model, NCAR physics adapted too Parallelized within WRF infrastructure

Hybrid and Eta Coordinates Ptop = 0 Pressure domain =0 Sigma domain ground =1 MSL =1

WRF Modeling System WRF Software Infrastructure Obs Data, Analyses Dynamic Cores Static Initialization Mass Core 3 DVAR Data Assimilation NMM Core … Standard Physics Interface Physics Packages Post Processors, Verification

WRF Hierarchical Software Architecture • Top-level “Driver” layer – Isolates computer architecture concerns – Manages execution over multiple nested domains – Provides top level control over parallelism • patch-decomposition Driver Layer wrf • inter-processor communication • shared-memory parallelism initial_config alloc_and_configure – Controls Input/Output • “Mediation” Layer Mediation Layer init_domain integrate solve_interface solve – – Model Layer Performs actual model computations Tile-callable Scientists insulated from parallelism General, fully reusable filter physics scalars recouple advance w advance uv decouple • Low-Level “Model” layer big_step prep – Specific calls to parallel mechanisms

• The National Weather Service dropped Eta in 2006 as the NAM (North American Mesoscale) run and replaced it with WRF NMM. • The Air Force uses WRF ARW. • Most universities use WRF ARW

WRF-NMM • Same domain as Eta • Sixty levels like Eta • Essentially same physics as ETA • Much better in terrain…doesn’t share the eta’s problems. • Clearly inferior synoptic initialization and synoptic forecast than GFS

NMM WRF NAM NMM upgrades December 2008, include • GDAS (GFS analysis) as initial first guess. use of global analysis (GDAS) for first guess at t-12 hour (the start of the analysis cycle) improves the evolution of synoptic scale features in the new NAM-WRF. This is found consistently throughout the 84 -hour forecast. • Improved physics higher resolution snow analysis and changes to snow impact on surface energy budget, increased absorptivity of model clouds

NMM • Generally inferior to GFS

Rapid Update Cycle-RUC

RUC • A major issue is how to assimilate and use the rapidly increasing array of offtime or continuous observations (not a 00 and 12 UTC world anymore! • Want very good analyses and very good short-term forecasts (1 -3 -6 hr) • The RUC ingests and assimilates data hourly, and then makes short-term forecasts • Uses the MAPS mesoscale model…which uses a hybrid sigma/isentropic vertical coordinate • Resolution: 13 km and 50 levels

13 km RUC Terrain elevation - 100 m interval Improvements expected from 13 km RUC - Improved near-surface forecasts - Improved precipitation forecasts - Better cloud/icing depiction - Improved frontal/turbulence forecasts NCEP computer upgrade allows RUC 13 to run in same time as current RUC 20

Data Type ~Number Freq. -------------------------Rawinsonde 80 /12 h NOAA profilers 30 / 1 h VAD winds 110 -130 / 1 h Aircraft (V, temp) 1400 -4500 / 1 h Surface/METAR 1500 -1700 / 1 h Buoy/ship 100 -150 / 1 h GOES precip water 1500 -3000 / 1 h GOES cloud winds 1000 -2500 / 1 h GOES cloud-top pres 10 km res / 1 h SSM/I precip water 1000 -4000 / 6 h -------------------------GPS precip water ~300 / 1 h Mesonet ~5000 / 1 h METAR-cloud-vis-wx ~1500 / 1 h ------------------------- Cloud analysis variables NCEP RUC 20 operational Observations used in RUC 13 (at NCEP June 2005)

RUC History – NCEP (NMC) implementations 1994 - First operational implementation of RUC - 60 km resolution, 3 -h cycle 1998 – 40 km resolution, 1 -h cycle, - cloud physics, land-sfc model 2002 – 20 km resolution - addition of GOES cloud data in assimilation 2003 – Change to 3 d. VAR analysis from previous OI (April) 2004 – Vertical advection, land use (April) PBL-depth for surface assimilation (September) 2005 – 13 km resolution, new obs, new model physics (June) 2007 – WRF-based Rapid Refresh w/ GSI to replace RUC

20 km RUC 13 km RUC Soil moisture – 22 z - 21 Feb 2005 Dark blue = water More detailed coastline with 13 km resolution

WRF RUC • A new version of RUC has been developed, but not yet operational that uses the WRF model instead of the MAPS model.