Summary

Model type: Dynamical - axisymmetric, intensity only
Model timeliness: Early model
Model status: Experimental (not part of NHC operational suite, provided to JTWC)
Basins run in: All basins
ATCF TECH identifiers: Deterministic run: CHIP
Individual ensemble members: CHPn, where n = 2 through 7
Ensemble mean: CHPM
Forecast period: 0 to 120 hours
Included in TCGP: Yes - all basins: early intensity plots (AL, EP, CP), late intensity plots (WP, IO, SH)
Domain:  
Vertical coordinate:  
Grid: Potential radius coordinate
Cumulus parameterization: quasi-equilibrium (fixed relaxation time scale) 
Microphysics parameterization:  
Boundary layer parameterization:  
Radiation parameterization:  
Ocean coupling: 1D (mixing only)
Initialization method: Mid-level entropy is adjusted to match observed storm intensity from genesis up until beginning of forecast period
Initial and boundary conditions: GFS fields; GFS analyzed SST; mixed layer depth and sub-mixed layer thermal stratification from Levitus monthly climatology
Primary contact: Professor Kerry Emanuel, Massachusetts Institute of Technology (MIT)
Model website: http://wind.mit.edu/~emanuel/storm.html
Full documentation: http://wind.mit.edu/~emanuel/CHIPS.pdf

Brief Technical Description

The Coupled Hurricane Intensity Prediction System (CHIPS) is a relatively simple two-dimensional (radius-height plane) dynamical model. The use of a potential radius coordinate allows the model to provide high resolution in the eyewall region, while giving lower resolution in the outer part of the model domain. Convection is parameterized as a quasi-equilibrium balance between the increase of entropy due to surface fluxes, radial transport, and the downward transport of low entropy air from mid-levels. Both the initial and outer boundary value of the mid-level moist entropy is provided from NCEP operational analyses. Since the model is axisymmetric, it cannot represent environmental vertical wind shear, so the effect of vertical shear is parameterized via the model's convective fluxes (more shear produces more ventilation of the storm, causing weakening). The model is coupled to 1D ocean models that are strung out along the storm's predicted path.

Initialization Method

The model is started from the beginning of the storm history, with the mid-level entropy being adjusted to keep the intensity as close as possible to the observed value up until the beginning of the forecast. The potential intensity is computed from the GFS model's full vertical column profiles of temperature and humidity, but lagged by 5 days so as to remove any influence of the storm. 

CHIPS Ensembles

CHIPS is also run as a 7-member ensemble. The ensemble is perturbed as follows:

  • Member 1 - CHIP: control (same as deterministic run)
  • Member 2 - CHP2: initial intensity enhanced by 3 m/s in previous 24-hr period (ramped up)
  • Member 3 - CHP3: initial intensity decreased by 3 m/s in previous 24-hr period (ramped down)
  • Member 4 - CHP4: initial intensity same as control, but the intensity 12 hours previous is enhanced by 1.5 m/s to produce a negative intensification anomaly at the initialization time
  • Member 5 - CHP5: initial intensity same as control, but the intensity 12 hours previous is decreased by 1.5 m/s to produce a positive intensification anomaly at the initialization time
  • Member 6 - CHP6: initial intensity enhanced by 3 m/s in previous 24-hr period (ramped up) with vertical wind shear set to zero at all times -- meant to provide a plausible upper bound for the intensity forecast
  • Member 7 - CHP7: initial intensity decreased by 3 m/s in previous 24-hr period (ramped down) with vertical wind shear enhanced by 5 m/s -- meant to provide a plausible lower bound for the intensity forecast

References

Emanuel, K., C. DesAutels, C. Holloway, and R. Korty, 2004: Environmental control of tropical cyclone intensity. J. Atmos. Sci., 61, 843-858. (link to article)

 

This page was last updated 19 December 2012 by Jonathan Vigh (NCAR/RAL). It has been reviewed by Professor Kerry Emanuel.