Background (from NASA)
Hurricanes (in the Atlantic) or typhoons (in the Pacific) are Earth’s strongest cyclones. A tropical storm becomes a hurricane when sustained wind speeds reach 64 knots (74 mph). Accurate predictions of their tracks and intensities can save lives and minimize property loss.
While hurricane tracks can be predicted with fairly good accuracy, the prediction of hurricane intensity remained a challenge to operational forecasting. This page is based on the article by Kelley and Stout (2004).
Condensational heating associated with precipitation is a major driving force of atmospheric circulation. Back in the fifties, it was proposed that most of the heating of the upper troposphere in the Tropics are concentrated in a few towering cumulonimbus clouds rather than large scale uplifting (Riehl and Malkus, 1958 and Malkus, 1959). The “hot towers” which refers to the tall cumulonimbus, has been seen as one of the mechanisms by which the intensity of a tropical cyclone is maintained. Because of the size (1-5 km) and short duration (30 minute to 2 hours) of these hot towers, studies of these events have been limited to descriptive studies from aircraft observations, although a few have attempted to use the presence of hot towers in a predictive capacity.
Before TRMM, no data set exists that can show globally and definitively the presence of these hot towers in cyclone systems. Aircraft radar studies of individual storms lack global coverage. Global microwave or Infrared sensor observations do not provide the needed spatial resolution. With a ground resolution of 5 km, the TRMM Precipitation Radar provided the needed data set for examining the predictive value of hot towers in cyclone intensification.
After six years of operations, the TRMM satellite observed over 100 tropical cyclones. In most cases, the TRMM was able to sample the cyclones twice or three times. The best case is hurricane Isabel that made landfall in the mid-Atlantic regions in September 2003. TRMM was able to sample the hurricane six times during its life cycle.
Figure 3 shows the vertical structure of Isabel using PR through both its intensification and decay stages. These figures do show the presence of high convective towers in its intensifying stages.(Courtesy TSDIS)
To examine the “Hot tower” theory, Kelley and Stout (2004) collected data of these hot towers embedded in the hurricanes observed by TRMM. They classified the hurricane stages as intensifying and non-intensifying hurricanes.
Figure 4 (after Kelley and Stout, 2004) shows the vertical distribution of the tallest rain pixel in the eyewalls of tropical cyclones as observed by TRMM PR. The red and black lines show the distribution for intensifying and non-intensifying cyclones. There are distinct double peaks of the tallest convective PR pixel for the intensifying cyclone. After careful sensitivity and statistical analysis, they concluded that intensifying cyclones are more than twice likely to have a convective tower in their eye-wall than non-intensifying cyclones.
The data used in their analysis are Precipitation Radar (PR) Data
The precipitation radar (PR) was developed by CRL and NASDA in Japan. The instrument is a 128-element active phased array system, operating at 13.8 GHz. The nadir footprint of PR is 4.3 km, with a vertical resolution of 250m. The minimum radar reflectivity factor is about 18 dBZ, corresponding to a rain rate of about 0.5 mm/hour. It obtains unique rainfall information by its 215-km cross-track scan through nadir.
TRMM 2A25 contains vertical ranfall rate profiles for one orbit. Also provided are: attenuation corrected Z profiles, parameters of Z-R relation (the relation between Z and rainfall rate), integrated rainfall rate for each ray, range bin numbers of rain layer boundaries, and many intermediate parameters.
A granule of TRMM 2A25 consists of metadata, clutter flags, and swath data. See Readme for TRMM Product 2A25 for information on acquiring and accessing this data product.
Some of the images in Figures 1 and 3 are produced using the TSDIS OrbitViewer.