In this section we will look at the satellites we looked at and the techniques that were used to analyze information from them.
 

Info on GOES:

Spatial Coverage/Frequency:  GOES Soundings

                    Resolution 10 km
                 Scan Cycle:  One Hour
                 Two scans each - GOES-8 and GOES-9

                         Composite GOES-8/9 Hourly Sounder Coverage

 Source
 

 Goes Imagery Tutorial
 
 

The below figure and table show some of the satellite channels used in tropical meteorology.

 

Below is a figure of all of the sounder channels from GOES-8

CIMSS
 
 

SST from GOES

Sea Surface Temperature (SST) can derived from GOES by the following equation:

 SST = C0 + C1 * (BT(4)) + C2 * (BT(4) - BT(5))

where BT is the equivalent black-body (or 'brightness') temperature for the indicated band and the C's are the empirically determined coefficients.

The coefficients can be found from the ocean radiance and from SST measurements from buoys. The radiances are found between 10.2- 11.2 micrometers and between 11.5-12.5 micrometers. The FOV resolution for GOES SST is 4 km.

 For more info about GOES SST data click here.

 GOES SST Algorithm
 

Deriving lower-level and upper-level winds from satellite images

Experimental GOES high-density multispectral winds have been produced by the University of Wisconsin's Cooperative Institute for Meteorological Satellite Studies (CIMSS) since 1996.   CIMSS uses an automated cloud-drift wind system which tracks clouds and water vapor in successive satellite images.  The system can derive winds at different levels of the troposphere using multispectral imagery utilizing 3 successive infra-red window channels.  Low level winds are track cumulus clouds by utilizing high-resolution imagery at a wavelength of 0.6 micrometers.  Upper level winds are found by following cirrus clouds by using 11 micrometer infrared images.

Trackable clouds are selected by the system and are assigned heights by several processes.  Clouds that can be tracked are given heights when the system finds the best match between the radiance and a temperature profile of the region.  Then the computer searches successive satellite images, with 15-30 minute intervals, for the same cloud.  Track and speed vectors are found when the system can successfully track the cloud after three satellite images.

For clear regions, satellite-dervived winds can be found by using the 6.7, 7.0, and 7.3 micrometer water vapor channels.  These can provide additional coverage in the middle (350-550mb) and upper (150-350mb) level regions of the troposphere.  Once again three successive images are used to track water vapor features and they are assigned heights by looking at brightness temperatures (Soden et al. 2000).
For more info on satellite-derived winds click here.

Below is an image of targeted clouds for use in deriving winds from satellites

 CIMSS
 

Hurricane Models
 

Models

The Primitive Equations
 
 
 
 

Geophysical Fluid Dynamics Lab (GFDL) Hurricane Forecast Model

    Description:

• Operational hurricane forecast model since 1995
• Limited area baroclinic model.
• Solves primitive equations by utilizing a finite difference technique with 18 levels.
• Multiply nested grid with two outer meshes nested within an outer mesh
• Initial and lateral boundary conditions from NCEP global analysis and hurricane analysis from NHC.
• Contains a synthetic vortex to represent a hurricane-scale circulation near the center.
• Model integrated out to 72 hours for both intensity and track.

Assimilating GOES wind data into GFDL Model by                                                     Three-Dimensional Optimum Interpolation (3DOI) (Soden et al. 2000)

The first step used by Soden et al. to assimilate the GOES winds data into the GFDL model is to determine the increment field which is the difference between the GOES winds and NCEP's "first guess" synoptic analysis of winds.  Since there ar several thousand observations at the same time, the data is reduced into a 1 degree by 1 degree "super-observation" gridpoint.  This reduces the volume of observations by a factor of four.  By this 30 observations used at each gridpoint.  Each gridpoint is then weighted according to the anticipated error for the GOES winds data and NCEP's "first guess" wind data.  GOES winds are assumed to have a signal to noise ratio of 8:1.  Water vapor winds below 300mb are left out of the analysis as the quality control because of the difficulty in finding the thickness of the moisture layer in this region.  More in depth details of this method can be found in Theibaux and Pedder (1987), Daley (1991), Soden et al. (2000), and others.
 
 
 
 

TRMM Satellite Description

• a suite of sensors including one passive TMI (TRMM Microwave Imager), one VIS/IR scanner and one Precipitation Radar
• aims to produce best rainfall estimate from space
• based on algorithms that combine measurements from all sensors
• measure precipitation over tropical regions using passive and active microwave instruments
• 10.7 GHz

shows instantaneous rain rate at the surface in mm per hour

 For more info on TRMM click here.
 
 
 

The TOPEX Satellite

• launched into Earth orbit in August of 1992.
• provides early warnings of El Nino and La Nina
• cooperation between the U.S. and France
• orbits the Earth 830 miles above the surface
• measures the sea level along the same path every 10 days

uses dual frequency altimeter developed by NASA and CNES

Ocean Upper Layer Thickness and Hurricane Heat Potential from TOPEX Measurements

For info about deriving upper layer and hurricane heat potential click here.
 
 

ERBE

Outgoing Longwave Radiation

 For info on ERBE click here.
 
 
 

 Results                                                                       Back to Main Page