HURRICANE
MODELS INFORMATION
Welcome to the Hurricane Models Page
of Hurricane Alley.
This page contains information of interest to
those who want to know what computer models are being used for
forecasting and a layman's explanation of what those models do
and the basis for their design and operation. This page lists
and describes the "movement", "intensity", and "surge" models.
STATISTICAL
MODELS
The statistical models start with
the information as to where the storm is located and the time of
year of the observation. The program will then search the
available database for other storms in the same location at the
same time of year. The forecast is then based upon the
history of those storms, what they did at the same time of year
from the same location. The program is not provided with
any information concerning current weather factors that may
influence the system being forecasted. This means that
there could be potentially major influences upon the particular
storm in question that would override the "historical"
perspective.
CLIPER
(CLImatology
and PERsistence)...
The predictors for
CLIPER include the initial latitude
and longitude of the storm, the components of the storm motion
vector, or which direction it is moving, the day of the year,
and the initial storm intensity. The
CLIPER
forecasts are used to normalize the output from the other
forecast models and as benchmark for tracking forecasting model
skill. This is the simple type of model that most hurricane
tracking software programs offer. It consists of a set of
equations that separately predict future zonal (east-west) and
meridional (north-south) movements of a tropical cyclone at
12-hr intervals out to 72 hr. The predictors include the current
and previous 12-hr position, the current and 12-hr previous
storm motion, the day of the year, and the maximum surface wind.
The initial motion of the storm (persistence) is the most
important predictor for this model. The skill of more complex
forecast models is often compared to that of
CLIPER. Any
model that cannot demonstrate significant skill over
CLIPER's
combination of climatology and persistence is discarded.
NHC98
This is the sixth in a series of models that is a
combination statistical and dynamical models that use the output
from
CLIPER, in
combination with vertically averaged winds through the
atmosphere and upper atmosphere air pressures from the AVN
(Aviation) run of the
MRF (Medium
Range
Forecast)
model as predictors. In NHC98, storms are stratified based
on their latitude and their current motion, with different
equations used for westward and eastward-moving storms. This
stratification is used to account for the observation that
storms within the easterlies tend to move to the right of the
steering flow, while storms within the westerlies tend to move
to the left of the steering flow. South Zone equations are used
for storms south of 15oN, and for storms between 15oN
and 25oN that are moving to the west or northwest.
North Zone equations are used for storms north of 25oN,
and for storms between 15oN and 25oN that
are moving to the north or northeast. NHC98 is run four times
per day. The primary synoptic time NHC98 forecasts (0000 and
1200 UTC) are based on the six hr-old aviation (AVN) run of the
NCEP global spectral model. A special version, NHC98-LATE, is
run at the primary synoptic times using forecasts from the
current AVN model run and is available several hours after
NHC98.
DYNAMICAL
MODELS
The dynamical models, unlike the statistical models, disregard
history altogether. They use as much information as
possible concerning the storm itself and the conditions
surrounding the storm. These models will use as much
"real-time" information as they can digest. The dynamical
models employ the basic laws of physics as they apply to the
atmosphere to predict the future course of the storm.
These models start with the six (6) basic equations concerning
these physical laws as they apply to the atmosphere. There
are three (3)
hydrodynamic
equations which use Newton's second law of motion to find the
horizontal and vertical motions of air caused by air pressure
differences, gravity, friction, and the earth's rotation.
There are two (2)
thermodynamic
equations which calculate changes in temperature caused by by
the evaporation of water into water vapor, the vapor condensing
into liquid, and so on. The final equation is known as the
continuity equation.
This equation attempts to account for volume of air going into
or coming out of a specified area. One form of a the
dynamical models is the
barotropic
model. This model moves weather systems in from one
location to another using horizontal winds only. In an
undisturbed, no major systems, type of atmosphere that is
usually found in the tropics devoid of a storm, this process
works extremely well. But, as the storm develops cold or
warm air moving across lines of equal air pressure, or isobars,
is the dominant feature for any developing storm.
Therefore, the barotropic becomes the least valuable. When
the lines of equal temperature and equal pressure cross each
other, this then becomes a
baroclinic
type atmosphere. The simplest type of dynamical model sets
up a three-dimensional grid of the atmosphere of isolated points
on the earth's surface. Observational readings are then
taken which include winds, air pressure, humidity and
temperature. These readings are then fed into the computer
and the "model" will then create a forecast of future movement
based on the output from the interaction of the storm with these
atmospheric conditions at the selected grid points.
Obviously, the more grid point, the more data, the more accurate
the forecast. As well, if that data is taken at various
levels in the atmosphere the accuracy grows even more. The
dynamical models then take all this information and process it
to produce the forecasts. The larger the grid the more
reliable the forecast. These grid points are also "nested"
together so that the finer the nesting the more the reliability
of the forecast is increased. The major problem with this
process is the computing power and time it takes to process.
An example of the problem, if the fineness of the grid points or
mesh for an area the size of the globe is used with the amount
of data used for the most detailed of the models the result
would be the most accurate forecast possible, but it would take
7 days, or one week, to produce a 24 hour forecast. Not
bad if you want to wait a week and see what the forecast was for
the first day of last week. Computers can only do so much.
So we live with the inherent forecast inaccuracies so that we
can gain the amount of accuracy that we can truly expect to get
given the current state of the processing power of today's
computers.
AVN
(Aviation)...
The AVN
or Aviation model is run by the NCEP, or National Centers for
Environmental Prediction MRF (Medium Range Forecast) model.
The MRF is a 28-level of atmosphere global model. That
means it is run using readings from 28 levels of the atmosphere
over the entire globe. It includes parameterizations for
convective (thunderstorm), radiative (returned sunlight), and
boundary layer processes. For tropical cyclone forecasts,
it uses synthetic observations of s storm's core that are
constructed from an estimate of the central pressure, the value
and radius of maximum low-level winds, the radii of 34-knot
winds, and the radius and pressure of the outermost closed
isobar. The synthetic are included at 50 sites within
approximately 200 nautical miles of the storm's center from the
surface to the maximum level of the storm's circulation, which
in most cases is somewhere near 30,000 feet. An automated
tracking algorithm provides a forecast track out to 72 hours.
The AVN/MRF
model differs from the GFDL Hurricane Model (GHM) model in that
it has a global domain, and the fields within the model are
represented by a set of mathematical functions rather than
values at discreet grid points. The forecast equations are
solved for the coefficients of the mathematical functions.
BAM
(Beta
and Advection
Model)...
This model follows a trajectory from the Aviation run
of the MRF model to provide a track forecast. This model
incorporates a correction known as the "Beta
Effect". This is used to
account for the fact that the Coriolis force resulting from the
rotation of the earth is greater toward the poles, so the winds
on the northern side of the storm in the Northern Hemisphere
tropical cyclone are turned more than those on its southern
side. If no other winds were steering a tropical cyclone
the "Beta Effect"
would cause a westward-headed storm to drift toward the north in
the Northern Hemisphere, and toward the south in the Southern
Hemisphere. There are three (3) versions of the
BAM...
1..
BAMS -
the BAM
Shallow,
this version averages winds from 5,000 to 10,000 feet (850 - 700
mb)
2..
BAMM -
the BAM
Medium,
this version averages winds from 10,000 to 24,500 feet (850 -
400 mb)
3..
BAMD -
the BAM
Deep,
this version averages winds from 24,500 to up to 47,000 feet
(850 - 200 mb)
For a weak hurricane without a well-developed eye wall
extending deep into the atmosphere, or for a tropical storm, the
shallow version of the model may work well, because storms of
this nature tend to be steered by low-level winds. As the
storm grows stronger and the eye wall gets deeper the deeper
versions become more accurate, for these types of storms are
steered more by the winds in the upper-level. If the
forecast from the three versions is similar the forecaster can
then assume that the storm may go as predicted, but, if the
version vary by a great deal, then the forecaster has less
confidence in the track predicted. The large differences
can also point to wind shear in the atmosphere, which could
affect the intensity forecast as well.
GFDL
(Geophysical
Fluid
Dynamics
Laboratory)...
This model is a limited area baroclinic model. It
was developed specifically for hurricane prediction. It
uses 3 nested grids across 18 levels. The two (2) inner
grids move to follow the storm. The resolution of the
inner grid is
1/6° of latitude.
The GFDL
model includes radiative, convective, and boundary layer
parameterizations. It has a specialized method for
initializing the center of the storm's circulation. The
initial and boundary conditions are obtained form the Aviation
run of the MRF model. The representation of the
storm's circulation in the global analysis is replaced with the
sum of an environmental flow and an idealized vortex. This
idealized vortex is based on the results of a few actual
parameters of the observed storm including the maximum wind, the
radius of the maximum wind, and the outer wind radii. The
environmental flow is the global analysis modified by a
filtering process that the removes the storm and its circulation
itself from the environment.
GHM
- The GFDL
Multiply-Nested Moveable Mesh Hurricane
Model
The GHM
is a dynamical baroclinic track prediction model. The model also
produces experimental forecasts of hurricane intensity and wind
swath maps that show the distribution of predicted maximum
surface and boundary layer winds. The
GHM was
developed by NOAA's Geophysical Fluid Dynamics Laboratory at
Princeton University. The GHM
is a triply nested, moveable mesh primitive equation model
formulated in latitude, longitude, and sigma coordinates. The
grid configuration of the GHM
was modified on May 21, 2002. Under the new two-nest grid
configuration the region covered by 1/6 degree resolution was
increased from five to 11 degrees. This area corresponds to the
region previously covered by the 1/3 degree resolution middle
grid. In addition, the resolution of the outermost grid was
changed from one degree to 1/2 degree. The model has 18 vertical
levels. The storm is centered in the middle of the finest grid
at the start of an integration. Lateral boundary conditions are
obtained from the AVN runs of the NCEP global spectral model.
There is two-way interaction between the grids, i.e., features
that form during an integration on the innermost grid are passed
to the outer grids, and vice-versa. The
GHM forecasts
are available about five hours after the primary and
intermediate synoptic times (0000, 0600, 1200 and 1800 UTC). To
overcome this shortcoming, the Tropical Prediction Center has
developed an interpolation technique to transpose the forecast
from the previous run to the current storm position. This
procedure is used for all the "late" models (i.e., those that
depend on the AVN model for their lateral boundary conditions).
The
GUNS Ensemble
- An Average of the GFDL,
UKMET
Office and NOGAPS
Models
James Goerss of the Naval Research Laboratory in Monterey,
California, has demonstrated that a simple consensus of the
GFDL, UKMET and NOGAPS models was about 20% more accurate at 24,
48 and 72 hrs than the best of individual models. The National
Hurricane Center confirmed his results and dubbed the ensemble "GUNS,"
using the initials of the three models. Consensus forecasts,
on average,
are often more accurate than the forecasts from individual
models, and the spread of an ensemble has potential use as a
measure of confidence in the forecast.
LBAR
(Limited
area BARotropic)...
This model is a two-dimensional track prediction model that is
initialized with vertically-averaged winds and upper atmospheric
air pressures from the Aviation run of the
MRF
global model. An idealized symmetric vortex is added to
the global analysis to represent the storm's circulation.
The boundary conditions are obtained from the global model
forecast. LBAR
is the NHC's implementation of the
GFDL
VICAR model. (VICBAR
stands for Vic Ooyama's Barotropic model.) The storm environment
domain analysis is produced with a two-dimensional spectral
application of finite element representation, using all
available data (rawindsondes, cloud drift winds, aircraft
observations, etc.), with the NCEP
global model analysis used as a low level background field. The
vortex domain analysis consists of synthetic observations
representing storm circulation and current storm motion. The
vortex is prescribed to be the same size and intensity in all
directions (axisymmetric), with winds increasing linearly from
the center to the radius of maximum winds. Wind speeds beyond
the radius of maximum winds are prescribed to decrease
exponentially to the edge of the storm. In the event of multiple
tropical cyclones, synthetic vortices are included for each
storm. The simplicity of barotropic models means they can be run
quickly on inexpensive computers. In the
LBAR
prediction model, the shallow water equations are solved on a
series of nested grid meshes on a Mercator projection. The inner
meshes move to remain centered on the storm, while the outer
mesh is fixed geographically. Time-dependent boundary conditions
from the AVN model run are applied outward from a transition
zone between 1500 and 2500 kin. LBAR runs on a 6-hr forecast
cycle and produces forecasts out to 72 hr.
NOGAPS
(Naval
Operational
Global
Atmospheric
Prediction
System)...
This model has 18 levels and is global in scope.
It uses parameterizations of physical processes and a bogussing,
or faking, scheme for a tropical cyclone. In general
terms, the bogussing scheme of the
NOGAPS is similar to that used by
the MRF. In this process synthetic observations that
represent the storm's circulation are added to the data
assimilation system. As with the MRF scheme, the
observations are built from a symmetric vortex and the sum of
the environmental flow.
UKMET
(United
Kingdom
Meteorological
Office)...
Like the NOGAPS and MRF models, the
UKMET
includes extensive physical parameterizations and a tropical
cyclone bogussing system.
INTENSITY MODELS
SHIPS
Statistical
Hurricane
Intensity
Prediction
Scheme Model
The SHIPS
model is a statistical-dynamic intensity prediction model. This model was
developed using standard multiple regression techniques with climatological,
persistence, and synoptic predictors. Estimates of future storm intensity are
made for 12-hr periods out to 72 hr. The SHIPS
equations were initially developed using data from 49 storms during the period
1982-1992 that were at least 30 nautical miles from land. The equations have
been updated using data from 1989 through the 1996 seasons. The primary
predictors used in the equation are (1)
Intensification potential (the difference between the
current storm intensity and an estimate of the Maximum Possible Storm Intensity
determined from the sea surface temperature);
(2) the vertical shear of the horizontal wind in
the 850 - 200 mb layer; (3)
persistence (intensity change in previous 12 hrs);
(4) average
200 mb temperature; (5)
average 200 mb east wind component;
(6) average 850 mb
vorticity; (7)
day of the year; (8)
and the flux convergence of eddy angular momentum
evaluated at 200 mb. Vertical wind shear is
evaluated for the 850 - 200 mb layer because most satellite cloud track winds
are assigned to those levels. The flux convergence of angular momentum tends to
be large whenever a storm approaches an upper-level trough and the upper level
winds over the storm are primarily from south to north. The sea surface
temperature, the 200 mb temperature and wind components, the 850 mb vorticity
and the vertical shear are averaged along the forecast track of the storm
derived from the VICBAR track guidance model. The other predictors are evaluated
from synoptic fields. An 11-level, no-physics version of a limited-area
baroclinic model, with boundary forcing from the AVN, is now used to produce the
forecast synoptic fields. Since the SHIPS
equations were developed using data from storms that were over water, the SHIPS
intensity forecasts are not valid for storms near the coast. In 2000 a new
version of the model, called Decay SHIP
(DSHP), was
introduced. The DSHP
is identical to the SHIPS
model except, if the cyclone is forecast to cross land, the intensity is reduced
accordingly.
STORM SURGE MODELS
SLOSH - The
Sea
Lake and
Overland
Surges from
Hurricanes Model
When hurricane warnings contain the range of
expected peak storm surge heights within the hurricane-warning area,
the surge information is often based on the
SLOSH model. The
dynamical SLOSH
model computes the water height over a geographical area or basin.
Computations have been run for a number of basins covering most of
the Atlantic and Gulf Coasts of the U.S. and the offshore islands.
The typical SLOSH
grid contains over 500 points located on lines extending radially
from a common basin center. The distance between grid points ranges
from 0.5 km near the center (where surge water heights are of more
interest), to 7.7 km in the deep water at the edge of the grid.
Bathymetric and topographic map data are used to determine a water
depth or terrain height for each grid point. The model consists of a
set of equations derived from the Newtonian equations of motion and
the continuity equation applied to a rotating fluid with a free
surface. The equations are integrated from the sea floor to the sea
surface. The coastline is represented as a physical boundary within
the model domain. Subgrid-scale water features (cuts, chokes, sills
and channels), and vertical obstructions (levees, roads, spoil banks,
etc.) can be parameterized within the model. Astronomical tides,
rainfall, river flow, and wind-driven waves have not been
incorporated into the model. The primary use of the
SLOSH model is to
define flood-prone areas for evacuation planning. The flood areas are
determined by compositing the model surge values from 200-300
hypothetical hurricanes. Separate composite flood maps are produced
for each of the five Saffir-Simpson hurricane categories. The
SLOSH model can also be run using
forecast track and intensity data for an actual storm as it makes
landfall. The model is highly responsive to the point of landfall,
however.
