The PSDI Tropical Cyclone Formation Probability product is currently operational and has been running at SAB throughout the 2006 Atlantic & E. Pacific hurricane seasons http://www.ssd.noaa.gov/PS/TROP/genesis.html.  Verification for the 2006 season is currently underway.  Data collection and quality control have been completed for the PSDI Tropical Cyclone Formation Probability project extension, which expands the original product domain of the Atlantic and E. Pacific to include the Central and Western N. Pacific.  Complete Best Track data for the Central & Western N. Pacific basins from 1949-present was acquired in Fall 2006, from which monthly climatological tropical cyclone formation probabilities were developed.  Plots of this climatology can be found at http://rammb.cira.colostate.edu/projects/gparm/climatologicalprobabilities/index.html.  Water vapor imagery from the GMS-5, GOES-9 and MTSAT-1 satellites  over the Western N. Pacific from 2000-present and from the GOES-10 satellite over the Central N. Pacific have been obtained from various sources.  With data collection complete, product development has begun and first experimental tests are planned for the 2007 typhoon season. (A. Schumacher, M. DeMaria, J. Knaff)

Last quarter much progress has been made in the development of a tropical cyclone web page.  The purpose of this web site is to provide real-time information on tropical cyclones with particular emphasis on products that were developed by RAMMB.  Current products that are displayed for all active tropical cyclones 1) Current information about the storm, 2) the latest forecast, 3) earth fixed java loops of IR imagery that are part of the automated tropical cyclone IR image archive 4) 1km mercator remapped  visible and IR images from the suite of NOAA satellites and NASA aqua and terra, 5) a six hourly multi-platform satellite-only tropical cyclone wind analysis, 6)  Forecast tracks overlaid upon maps of ocean heat content from NOAA/AOML 7) Temperature and tangential wind radial-height cross-sections derived from AMSU temperature retrievals, 8) Time series of current intensity with AMSU-based intensity estimates overlaid, and 9) Comparisons of current intensity with intensities estimated from the original digital Dvorak methodology.   It is important to note that the web page is integrated with a MS Access database, which allows access to products created for past storms and that the web site is still under construction.  (J. Knaff)

Three new automated products have been created for the aforementioned web page this quarter.  Two of these are derived from the CIRA/NESDIS AMSU-based intensity and structure estimates that are now run at NCEP Central Operations and disseminated via ftp. The first AMSU-based product is the azimuthal mean radial-height cross sections of temperature and derived tangential winds (assuming gradient balance) as shown in Figure 1.  The second a time series of warning intensity versus the AMSU derived intensities (Figure 2).  The final new product shows intensity estimates by the Digital Dvorak technique.  This product produces a current intensity estimate (CI) in terms of T-numbers and storm intensity (kt) using the latest position and motion vector to estimate the center location.  Figure 3 shows the graphic associated with this product.  At the top are comparisons of the Digital Dvorak intensity vs. the operational intensity estimate and at the bottom the raw T-numbers, a six-hour running mean of the raw T-numbers and the CI-number.  (J. Knaff)

Click on images to enlarge.

Figure 1:  An example of the radial-height cross sections of temperature and tangential wind for Typhoon Utor on 10 December at 09 UTC.

Figure 3: An example of the time series plots created from the Digital Dvorak Technique for Typhoon Utor.

Based upon research presented in Knaff et al. (2003) a new class of hurricanes with a very large and symmetric eyes and little precipitation outside of the eye was discovered. These storms, termed “annular hurricanes”, have intensity change characteristics that are different than the larger sample of storms in that they are more stable, and do not fluctuate in intensity as much as typical hurricanes. Although fairly rare (an average of about one per year in the Atlantic), the identification of these storms provides valuable input to the forecasters at the National Hurricane Center. An annular hurricane index that provides a quantitative method for real-time identification of these storms was developed and implemented on NCEP computer systems utilized by NHC, as part of their operational Statistical Hurricane Intensity Prediction Scheme. The index first identifies whether annular structure is possible, and, if so, provides an index value between 1 and 100 (100 is best) that determines how well the storm fits the annular structure based upon GOES imagery and supplemental information about the storm environment. The index was tested on all Atlantic and east Pacific tropical cyclones from the 2006 hurricane season, and will be run in real time beginning in 2007, or if there is a late season storm in the remainder of 2006. Figure 4 below shows an example of Hurricane Daniel from the 2006 eastern Pacific season that was identified by the index as having strong annular characteristics.

Click on image to enlarge.

Figure 4: GOES infrared imagery of Hurricane Daniel at 0000 UTC on 22 July 2006 when it was identified by the new annular hurricane index as having strong annular characteristics.

A paper discussing an updated AMSU-based intensity and structure estimation algorithm was published.   Previous work, in which Advanced Microwave Sounding Unit (AMSU) data from the Atlantic Ocean and east Pacific Ocean basins during 1999–2001 were used to provide objective estimates of 1-min maximum sustained surface winds, minimum sea level pressure, and the radii of 34-, 50-, and 64-kt (1 kt 0.5144 m s−1) winds in the northeast, southeast, southwest, and northwest quadrants of tropical cyclones, is updated to reflect larger datasets, improved statistical analysis techniques, and improved estimation through dependent variable transforms The developmental regression models resulted in greatly improved intensity and structure estimates.  These algorithms have been running since May 2006 at NCEP Central Operations and provide tropical cyclone intensity estimates for global tropical cyclones being monitored by (NOAA/TPC, NOAA/CPHC, and DOD/JTWC).  (J. Knaff, M. DeMaria)

Return to the current RAMM Branch Quarterly Report