2012 HFIP Demonstration | Regional Models

  • Overview
  • MMM and Albany-SUNY
  • NRL
  • GFDL
  • FSU
  • NESDIS/STAR and CIRA
  • UWIS
  • PSU
  • EMC

For the annual HFIP Demo, real-time forecasts are generated using experimental numerical weather prediction model configurations that attempt to provide improvements in tropical cyclone guidance forecasts. Each modeling group has extensive expertise in hurricane forecasting and is currently employing a broad spectrum of physics parameterizations, data assimilation techniques and coupling complexities. The groups running regional models were the National Center for Atmospheric Research's Mesoscale and Microscale Meteorology Division (MMM) and the University at Albany-State University of New York (Albany-SUNY), the Naval Research Laboratory (NRL), NOAA's Geophysical Fluid Dynamics Laboratory (GFDL), Florida State University (FSU), the Center for Satellite Applications and Research (STAR) of the National Environmental Satellite, Data and Information Service (NESDIS), the Cooperative Institute for Research in the Atmosphere (CIRA), University of Wisconsin (UWIS), Pennsylvania State University (PSU), and the Environmental Modeling Center (EMC) of NOAA's National Centers for Environmental Prediction. Basic information on each model configuration is provided under the tab with the corresponding abbreviated group identifier.

National Center for Atmospheric Research - Mesoscale and Microscale Meteorology Division (NCAR/MMM) and the University at Albany-State University of New York (Albany-SUNY)

ATCF IDs - AHW4 [Deterministic], AHnn (nn=01-15) [Ensemble]

Contact Person: Chris Davis, Ryan Torn, Wei Wang

AHW

Overview

The WRF model is designed to be a flexible, state-of-the-art, portable code that offers two dynamic solvers and numerous physics options. The Advanced Research WRF (ARW) solver developed at NCAR, utilizes the Arakawa C grid on several different projections and a terrain-following mass coordinate. The ARW also employs higher order numerics for advection and time integration. For the 2012 HFIP Demo, the MMM configuration of ARW uses the Lambert-Conformal projection with one static domain and two moving nested domains. For more detailed information on the ARW, please see Skamarock et al., 2008.

Domains

Horizontal

36 km (320 by 210) / 12 km (133 by 133) / 4 km (199 by 199)

Vertical

35 full levels with model top at 20 mb

Atmosphere

Initialization

Atmospheric fields for the retrospective forecasts are initialized with fields obtained from the Ensemble Kalman Filter (EnKF) method in a 6-hour cycling mode, starting at the beginning of August and cycling through the season. No vortex relocation or bogus is used. The land surface model was initialized with fields from the previous (6-h) GFS cycle.

Lateral Boundary Conditions

Lateral boundaries are taken from the previous (6-h) GFS cycle.

Physics

Cumulus Tiedtke
Microphysics WSM6
PBL YSU
Surface Layer Monin-Obukov
Land Surface NOAH
Radiation RRTMG (longwave and shortwave)

Ocean

Ocean mixed-layer model

Initialization

Hybrid Coordinate Ocean Model (HYCOM) upper ocean structure used to prescrive initial mixed-layer depth.

References

Davis, C., W. Wang, S. S. Chen, Y. S. Chen, K. Corbosiero, M. DeMaria, J. Dudhia, G. Holland, J. Klemp, J. Michalakes, H. Reeves, R. Rotunno, C. Snyder, and Q. N. Xiao, 2008: Prediction of landfalling hurricanes with the advanced hurricane WRF model. Mon. Wea. Rev., 136, 1990-2005.

Davis, C. A., W. Wang, J. Dudhia, and R. Torn, 2010: Does Increased Horizontal Resolution Improve Hurricane Wind Forecasts? Wea. Forecasting, 25, 1826-1841.

Torn, R. D., 2010: Performance of a mesoscale Ensemble Kalman Filter (EnKF) during the NOAA High-Resolution Hurricane Test. Mon. Wea. Rev., 138, 4375-4392.

Naval Research Laboratory (NRL)

ATCF ID - COTC

Contact Person: James Doyle, Richard Hodur, Hao Jin

Coupled Ocean/Atmosphere Mesoscale Prediction System – Tropical Cyclone (COAMPS-TC)

Overview

The Coupled Ocean-Atmosphere Mesoscale Prediction System (COAMPS) is the Navy high-resolution regional operational prediction system. COAMPS is developed by NRL and consists of data quality control, data assimilation, initialization, a non-hydrostatic atmospheric model and a hydrostatic ocean model (Hodur 1997). The Arakawa C grid is used for both the atmospheric and ocean models. The atmospheric model utilizes the sigma-z vertical coordinate and the ocean model uses the hybrid Sigma/z. A version of COAMPS has recently been developed which is dedicated to the prediction of tropical cyclones (COAMPS-TC). For the 2012 HFIP Demo, the model is run on a Mercator projection with one fixed coarse mesh domain and either two or three moving, two-way interactive nested domains.

Domains

Horizontal

45 km / 15 km / 5 km

Vertical

40 levels with model top at 31 km

Atmosphere

Initialization

Data assimilation cycle using synthetic observations. Force-restore — NOAH option is used for the land surface model.

Lateral Boundary Conditions

Boundary conditions are obtained from NOGAPS.

Physics

Cumulus Kain Fritsch
Microphysics NRL
PBL NRL 1.5 order closure
Surface Layer NRL / CBLAST
Land Surface Force-restore
Radiation Fu-Liou

Ocean

Constant SST and coupled ocean model

Initialization

NRL Coupled Ocean Data Assimilation (NCODA) with NRL Coastal Ocean Model (NCOM) when model is coupled. For retrospective test cases, the model uses constant SST (NCODA SST analysis).

National Oceanic and Atmospheric Administration - Geophysical Fluid Dynamics Laboratory (NOAA/GFDL)

ATCF IDs - GPnn (nn=00-15) [Ensemble], GPMN [Ensemble Mean]

Contact Person: Tim Marchok

Geophysical Fluid Dynamics Laboratory (GFDL) Ensemble

Description of Members

The 2012 Geophysical Fluid Dynamics Laboratory (GFDL) regional hurricane ensemble consisted of 16 members based on NCEP's 2012 implementation of the GFDL hurricane model. The initial and boundary conditions for half of the members were based on forecasts from NCEP's Global Forecast System (GFS) and the other half used mean fields from NCEP's Global Ensemble Forecast System (GEFS). A 'control forecast' was produced for each set of initial and boundary conditions, with the remaining seven members for each group generated by employing perturbations to the initial conditions targeting properties of the storm core region (vortex structure, moisture fields, and SST fields). A detailed description of each ensemble member, as well as the method used to compute the ensemble mean is found in Table 1.

Table 1: Summary table of final setup for the GFDL hurricane ensemble. Global model used for initial and boundary conditions are noted in the brackets.
ATCF ID Description
GP00 Control forecast [GFS deterministic] (same model as NCEP operational GFDL)
GP01 Unbogussed forecast using the 2012 control model [GFS deterministic]
GP02 Increase NHC-observed Vmax 10%, 34kt radii 25%, 50/64kt radii 40%, ROCI 25% [GFS deterministic]
GP03 Decrease NHC-observed Vmax 10%, 34kt radii 25%, 50/64kt radii 40%, ROCI 25% [GFS deterministic]
GP04 Modification to increase inner-core moisture by a max of 10% [GFS deterministic]
GP05 Modification to decrease inner-core moisture by a max of 10% [GFS deterministic]
GP06 Increase SSTs by a max of 1°C within the initial extent of the TC [GFS deterministic]
GP07 Decrease SSTs by a max of 2°C within the initial extent of the TC [GFS deterministic]
GP08 Control forecast [NCEP GEFS ensemble]
GP09 Unbogussed forecast using the 2012 control model [NCEP GEFS ensemble]
GP10 Increase NHC-observed Vmax 10%, 34kt radii 25%, 50/64kt radii 40%, ROCI 25% [GEFS ensemble]
GP11 Decrease NHC-observed Vmax 10%, 34kt radii 25%, 50/64kt radii 40%, ROCI 25% [GEFS ensemble]
GP12 Modification to increase inner-core moisture by a max of 10% [GEFS ensemble]
GP13 Modification to decrease inner-core moisture by a max of 10% [GEFS ensemble]
GP14 Increase SSTs by a max of 1°C within the initial extent of the TC [GEFS ensemble]
GP15 Decrease SSTs by a max of 2°C within the initial extent of the TC [GEFS ensemble]
GPMN Computation of the ensemble mean applied at each lead time where 40% of the members are available

Florida State University (FSU)

ATCF ID - MMEN

Contact Person: Anu Simon

Multimodel Superensemble

Members

Table 1: Summary table of real time forecast models included in multimodel superensemble for track.
ATCF ID Description
HWRF (H212) Hurricane WRF
COTC Coupled Ocean/Atmosphere Mesoscale Prediction System - Tropical Cyclone (COAMPS-TC)
AHW4 NCAR AHW
ARFS FSU ARW
SPC3 Statistical Prediction of Intensity from a Consensus Ensemble
FIM9 Flow-Following Finite-Volume Icosahedral Model
UWN8 8 km UW-NMS
GFDL GEOPHYSICAL FLUID DYNAMICS LABORATORY (NOAA/GFDL)
NGPS NOGAPS
ECMF (only for training data from 2011) ECMWF
AVNI GFS

Table 2: Summary table of real time forecast models included in multimodel superensemble for intensity.
ATCF ID Description
HWRF (H212) Hurricane WRF
COTC Coupled Ocean/Atmosphere Mesoscale Prediction System - Tropical Cyclone (COAMPS-TC)
AHW4 NCAR AHW
ARFS FSU ARW
SHIPS Statistical Hurricane Intensity Prediction Scheme
SPC3 Statistical Prediction of Intensity from a Consensus Ensemble
UWN8 8 km UW-NMS
GFDL GEOPHYSICAL FLUID DYNAMICS LABORATORY (NOAA/GFDL)
FIM9 Flow-Following Finite-Volume Icosahedral Model
NGPS NOGAPS
AVNI GFS

National Environmental Satellite, Data and Information Service (NESDIS)/Center for Satellite Applications and Research (STAR) and Cooperative Institute for Research in the Atmosphere (CIRA)

ATCF IDs - see below

Contact Person: Mark DeMaria

Statistical Prediction of Intensity from a Consensus Ensemble (SPICE)

ATCF IDs

The S denotes Statistical, the D means the DSHP intensity scheme is used, the L means the LGEM intensity scheme used, and the P stands for prediction, which corresponds to the model name--Statistical Prediction of Intensity from a Consensus Ensemble, and represents the weighted consensus of the DSHP and LGEM components. Each of the intensity schemes will be run based off of the various dynamical models and their consensuses listed below, but only SPC3 will be delivered. For example, SLHW would be the LGEM scheme run on the HWRF model fields.

DSHP LGEM Prediction
GFS SDGS SLGS SPGS
HWRF SDHW SLHW SPHW
GFDL SDGL SLGL SPGL
COAMPS-TC SDCT SLCT SPCT
NOGAPS SDNG SLNG SPNG
Consensus-Regional SDCR SLCR SPCR
Consensus-Global SDCG SLCG SPCG
Consensus-All SDCA SLCA SPCA
Consensus of first 3 (GFS, HWRF, GFDL) SDC3 SLC3 SPC3
Consensus-Weighted SDCW SLCW SPCW

References

http://ams.confex.com/ams/pdfpapers/167916.pdf

University of Wisconsin (UWIS)

ATCF ID - UWN8

Contact Person: Greg Tripoli and William Lewis

University of Wisconsin - Nonhydrostatic Modeling System (UW-NMS) v7b

Overview

The UW-NMS model is a nonhydrostatic research model, used primarily to investigate the interaction of convection with mesoscale and synoptic-scale weather phenomena. The UW-NMS utilizes the Arakawa C grid on a rotated spherical projection and a geopotential height vertical coordinate. For the 2012 HFIP Demo, the UW-NMS is configured with one static domain and one moving nested domain.

Domains

Horizontal

40 km (205 by 150) / 8 km (92 by 92)

Vertical

33 levels with model top at 24.2 km

Atmosphere

Initialization

GFS analysis with modified Kwon and Cheong (2010) bogus vortex using TC vitals input. The bogus initialization incorporates several bug fixes, and the Andreas sea-spray parameterization is switched off.

Land surface model is initialized with fields from GFS analysis.

Lateral Boundary Conditions

GFS forecast

Physics

Cumulus CMEE
Microphysics Tripoli/Flatau/Hashino
PBL 1.5 level closure
Surface Layer Monin-Obuknov Similarity
Land Surface NOAH
Radiation RRTM

Ocean

Cloupled ocean model

Initialization

SST from FNMOC GHRSST 10-km analysis

Pennsylvania State University (PSU)

ATCF IDs - APSU and ANPS

Contact Person: Fuqing Zhang

WRF v3.3.1

Overview

The WRF model is designed to be a flexible, state-of-the-art, portable code that offers two dynamic solvers and numerous physics options. The Advanced Research WRF (ARW) solver developed at NCAR, utilizes the Arakawa C grid on several different projections and a terrain-following mass coordinate. The ARW also employs higher order numerics for advection and time integration. For the 2012 HFIP Demo, the PSU configuration of ARW uses one static domain, which covers the Atlantic Ocean and North America, while the inner domains follow the vortex center. For more detailed information on the ARW, please see Skamarock et al., 2008.

Domains

Horizontal

27 km (379 by 244) / 9 km (304 by 304) / 3 km (304 by 304)

Vertical

43 full levels with model top at 50 mb

Atmosphere

Initialization

GFS analysis, conventional observations, NOAA P3 airborne radar obserations are used to generate the initial fields with PSU WRF-EnKF anlaysis system.

Lateral Boundary Conditions

GFS forecast

Physics

Cumulus Grell-Devenyi ensemble (27 km domain only)
Microphysics WSM 6-class graupel
PBL YSU
Surface Layer Monin-Obukov
Land Surface thermal diffusion
Radiation RRTM (longwave) / Dudhia (shortwave)
Ck, Cd formulation Garratt

Ocean

None

References

Skamarock, W. C., J. B. Klemp, J. Dudhia, D. O. Gill, D. M. Barker, M. G. Duda, X.-Y. Huang, W. Wang, J. G. Powers, 2008: A description of the Advanced Research WRF Version 3. NCAR Technical Note, NCAR/TN-475+STR [Available at Mesoscale and Microscale Meteorology Division, National Center for Atmospheric Research, Boulder, CO, USA].

National Centers for E nvironmental Prediction - Environmental Modeling Center (NCEP/EMC)

ATCF IDs - HHYC

Contact Person: Hyun-Sook Kim

Hurricane Weather Research and Forecast (HWRF) System coupled to HYCOM

Overview

HWRF atmosphere coupled to HYCOM, in replace of POM in the operational HWRF.