Skip Top Navigation
JCSDA Logo and banner

JCSDA News and Newsletters

Note: The March 2012 Newsletter was updated on April 20, 2012 with the following changes: inclusion of a complete list of sponsors for the ITOVS Conference and corrections to the caption of the figure in the Winds Workshop summary.

Improving Tropical Cyclone Forecasts with Water Vapor and Temperature Information from Satellites

from the March 2012 JCSDA Quarterly

chart: Fig 1. Hurricane Irene (2011) central sea level pressure (SLP) forecast

Fig 1. Hurricane Irene (2011) central sea level pressure (SLP) forecast RMSE for 0-h (analysis) and 6-h to 72-h forecasts. Every 6 hours between 06 UTC 23 and 00 UTC 25 August 2011 the data are assimilated with WRF/DART followed by a 72-hour forecast. The SLP RMSE is calculated from comparisons with the best track observations from the National Hurricane Center.

Under a grant from the JCSDA, scientists with the University of Wisconsin’s Cooperative Institute for Meteorological Satellite Studies (CIMSS) are examining the use of water vapor and temperature observations from satellites to improve path and intensity forecasts for tropical cyclones. Lack of good temperature and water vapor information appears to be a limiting factor for accurate predictions of these systems.

The high vertical resolution atmospheric temperature and moisture profiles from the Atmospheric InfraRed Sounder (AIRS) are used to initialize/analyze the development of a hurricane. The Weather Research and Forecast (WRF) model and Data Assimilation Research Testbed – DART (WRF/DART) developed by the National Center for Atmospheric Research (NCAR) are used to assimilate the AIRS data and generate the forecasts. By assimilating these sounding measurements, the representation of environmental conditions around the hurricane is more realistic, and thus the path and intensity forecasts should be improved. The hurricane predictions are examined with and without the satellite atmospheric temperature and moisture information. We have conducted lifecycle forecast experiments for hurricane Irene (2011). The assimilation time window is 1 hour (minus/plus 30 minutes) for AIRS. Every 6 hours between 06 UTC 23 and 00 UTC 25 August 2011, the data are assimilated with WRF/DART using 32 ensemble members, and 72-hour forecasts are performed after each assimilation. Two types of forecast sequences are conducted: a control run, which assimilates data from radiosondes, satellite cloud winds, aircraft, ships, and land surface stations; and an experimental run, which assimilates the same data as the control run plus AIRS single field-of-view (SFOV) soundings in clear skies.

The root mean square errors (RMSE) of the hurricane track and intensity forecasts are calculated using verification data from the best track observations of the National Hurricane Center. The total forecast number for each lead time is 256. Hurricane Irene (2011) intensity (central sea level pressure) forecast RMSE for 0-h (analysis) and 6-h to 72-h forecasts are shown in figure 1. Since the background fields for the first analysis are from the Global Forecast System forecasts, both control and AIRS runs show relatively large errors in sea level pressure in the early stages; these errors are gradually reduced as the model progresses in time. However, the AIRS soundings consistently lead to improvement of intensity forecasts during the process. Results of additional experiments with WRF/3DVAR were consistent with those from WRF/DART.

The impact of assimilating Moderate Resolution Imaging Spectroradiometer (MODIS) and Advanced Microwave Scanning Radiometer - Earth Observing System (AMSR-E) total precipitable water (TPW) on tropical cyclone analyses is also investigated. MODIS has a high spatial resolution of 5 km but has limited spatial coverage in clear skies; AMSR-E has a coarse spatial resolution of 21 km but has the advantage of coverage in cloudy skies. The assimilation experiments were conducted with the WRF/DART for Typhoon Sinlaku from 8 to 13 September 2008. Results show that both IR (MODIS) and microwave (AMSR-E) TPW measurements improve the track and intensity analysis when compared with the control run which assimilates radiosondes, satellite atmospheric motion vectors, QuickSCAT winds, COSMIC GPS-RO, ship and land surface observations. Microwave TPW provides a better analysis for intensity than IR alone, while combining IR and MW data generally provides a more positive impact on TC track and intensity analysis than that from either IR alone or MW alone (see results for intensity in fig. 2). The next generation of GOES is expected to improve tropical cyclone forecasting by providing critical high temporal and spatial TPW distributions, as well as improved satellite derived winds and cloud products.

chart: Typhoon Sinlaku sea level pressure analysis

Fig 2. Typhoon Sinlaku sea level pressure analysis for September 8 - 13, 2008 from control and TPW assimilation along with the observations.

In summary, these preliminary results are encouraging and suggest a positive impact of available satellite temperature and water vapor information for forecasting hurricane tracks and intensity.

Jun Li, Jinlong Li, and Jing Zheng, CIMSS, and Tim Schmit, STAR

Newsletters supplied as PDFs


 
 
Modified April 20, 2012 5:04 PM
NOAA | NWS | NASA
Navy Weather | Air Force Weather
Privacy |  Link & Product Disclaimers | Information Quality
Accessibility | JCSDA webmaster
Hosted for JCSDA at STAR
Level A conformance icon, W3C-WAI Web Content Accessibility Guidelines 1.0Valid HTML 4.01