NOAA Tropical Analysis and Forecast Branch

A database was established for the period 1970–99 to assess the threat to life in the contiguous United States and adjacent coastal waters from Atlantic tropical cyclones. Freshwater floods caused more than one-half of the 600 U.S. deaths directly associated with tropical cyclones or their remnants during that 30-year period. More than three-quarters of the victims under age 13 died in rain-induced floods. Most fatalities occurred in inland counties. Storm surge losses were significantly (but perhaps only temporarily) less than in previous periods of comparable length. This paper presents a statistical summary of the casualties, explores reasons for the losses, and reviews efforts to mitigate the threats.

The 1997 Atlantic hurricane season is summarized and the year’s tropical storms, hurricanes, and one subtropical storm are described. The tropical cyclones were relatively few in number, short lived, and weak compared to long-term climatology. Most systems originated outside the deep Tropics. Hurricane Danny was the only system to make landfall. It produced rainfall totals to near 1 m in southern Alabama and is blamed for five deaths. Hurricane Erika was responsible for the season’s two other fatalities, in the coastal waters of Puerto Rico.

Activity during the 2001 hurricane season was similar to that of the 2000 season. Fifteen tropical storms developed, with nine becoming hurricanes and four major hurricanes. Two tropical depressions failed to become tropical storms. Similarities to the 2000 season include overall activity much above climatological levels and most of the cyclones occurring over the open Atlantic north of 25°N. The overall “lateness” of the season was notable, with 11 named storms, including all the hurricanes, forming after 1 September. There were no hurricane landfalls in the United States for the second year in a row. However, the season's tropical cyclones were responsible for 93 deaths, including 41 from Tropical Storm Allison in the United States, and 48 from Hurricanes Iris and Michelle in the Caribbean.

Abstract The 2004 Atlantic hurricane season is summarized, and the year’s tropical and subtropical cyclones are described. Fifteen named storms, including six “major” hurricanes, developed in 2004. Overall activity was nearly two and a half times the long-term mean. The season was one of the most devastating on record, resulting in over 3100 deaths basinwide and record property damage in the United States.

Abstract The intensity issue of hurricanes is addressed in this paper using the angular momentum budget of a hurricane in storm-relative cylindrical coordinates and a scale-interaction approach. In the angular momentum budget in storm-relative coordinates, a large outer angular momentum of the hurricane is depleted continually along inflowing trajectories. This depletion occurs via surface and planetary boundary layer friction, model diffusion, and “cloud torques”; the latter is a principal contributor to the diminution of outer angular momentum. The eventual angular momentum of the parcel near the storm center determines the storm’s final intensity. The scale-interaction approach is the familiar energetics in the wavenumber domain where the eddy and zonal kinetic energy on the hurricane scale offer some insights on its intensity. Here, however, these are cast in storm-centered local cylindrical coordinates as a point of reference. The wavenumbers include azimuthally averaged wavenumber 0, principal hurricane-scale asymmetries (wavenumbers 1 and 2, determined from datasets) and other scales. The main questions asked here relate to the role of the individual cloud scales in supplying energy to the scales of the hurricane, thus contributing to its intensity. A principal finding is that cloud scales carry most of their variance, via organized convection, directly on the scales of the hurricane. The generation of available potential energy and the transformation of eddy kinetic energy from the cloud scale are in fact directly passed on to the hurricane scale by the vertical overturning processes on the hurricane scale. Less of the kinetic energy is generated on the scales of individual clouds that are of the order of a few kilometers. The other major components of the energetics are the kinetic-to-kinetic energy exchange and available potential-to-available potential energy exchange among different scales. These occur via triad interaction and were noted to be essentially downscale transfer, that is, a cascading process. It is the balance among these processes that seems to dictate the final intensity.

Numerical model experiments are conducted to address the previously unexplained anomalously high storm surge along the Florida coast of Apalachee Bay during Hurricane Dennis (2005). The 2–3 m surge observed during this storm cannot be obviously explained by the relatively weak local winds over this bay 275 km east of the storm center. Realistic and idealized numerical experiments demonstrate that the along‐shore winds to the east of the storm center built a high sea level anomaly along the coast which traveled northward to Apalachee Bay as a topographic Rossby wave. The wave was amplified as the storm moved nearly parallel to the shelf and at comparable speed to the wave phase speed. These results suggest that enlarging the domain of the storm surge forecasting models can improve the surge forecast for a storm moving along a similar track, and have now been applied to operational use.

The 2002 Atlantic hurricane season is summarized. Although the season's total of 12 named storms was above normal, many of these were weak and short-lived. Eight of the named cyclones made landfall in the United States, including Lili, the first hurricane to hit the United States in nearly 3 yr.

Abstract The 2003 Atlantic hurricane season is described. The season was very active, with 16 tropical storms, 7 of which became hurricanes. There were 49 deaths directly attributed to this year’s tropical cyclones.

An evaluation of the performance of the National Centers for Environmental Prediction Medium-Range Forecast Model was made for the large-scale tropical forecasts and hurricane track forecasts during the 1995 hurricane season. The assessment of the model was based on changes to the deep convection and planetary boundary layer parameterizations to determine their impact on some of the model deficiencies identified during previous hurricane seasons. Some of the deficiencies in the hurricane forecasts included a weakening of the storm circulation with time that seriously degraded the track forecasts. In the larger-scale forecasts, an upper-level easterly wind bias was identified in association with the failure of the model to maintain the midoceanic upper-tropical upper-tropospheric trough. An overall modest improvement is shown in the large-scale upper-level tropical winds from root-mean-square-error calculations. Within a diagnostic framework, an improved simulation of the midoceanic tropical trough has contributed to a better forecast of the upper-level westerly flow. In the hurricane forecasts, enhanced diabatic heating in the model vortex has significantly improved the vertical structure of the forecast storm. This is shown to contribute to a substantial improvement in the track forecasts.

Abstract The hurricane season of 2006 in the eastern North Pacific basin is summarized, and the individual tropical cyclones are described. Also, the official track and intensity forecasts of these cyclones are verified and evaluated. The 2006 eastern North Pacific season was an active one, in which 18 tropical storms formed. Of these, 10 became hurricanes and 5 became major hurricanes. A total of 2 hurricanes and 1 tropical depression made landfall in Mexico, causing 13 direct deaths in that country along with significant property damage. On average, the official track forecasts in the eastern Pacific for 2006 were quite skillful. No appreciable improvement in mean intensity forecasts was noted, however.

The 2002 eastern North Pacific hurricane season is summarized and the year's tropical cyclones are described. The season featured 12 named tropical storms, of which 6 became hurricanes. Five of the six hurricanes reached an intensity of 100 kt or higher. There were two landfalling cyclones, Tropical Storm Julio and Hurricane Kenna. Kenna, which made landfall near San Blas, Mexico, with winds of near 120 kt, was responsible for four deaths.

Abstract The 2005 eastern North Pacific hurricane season is summarized, the individual tropical cyclones are described, and official track and intensity forecasts are verified and evaluated. The season’s overall activity was, by most measures, below average. While a near-average 15 tropical storms formed, many of them were relatively weak and short-lived. Seven of these storms became hurricanes, but only one reached major hurricane status (an intensity of 100 kt or greater on the Saffir–Simpson hurricane scale) in the eastern North Pacific basin. One of the hurricanes, Adrian, approached Central America in May but weakened to a tropical depression prior to landfall. Adrian was the only eastern North Pacific tropical cyclone to make landfall during 2005, and it was directly responsible for one fatality.

This paper addresses the weakening of Hurricane Lili of October 2002 just before it made landfall in Louisiana. This hurricane weakened from a category 4 storm on October 3, 2002 at 0000 UTC to a category 1 storm on October 3, 2002 at 1300 UTC. This sudden drop in intensity has been a subject of considerable interest. In this paper we explore a forecast model diagnostic approach that explores the contribution to the hurricane intensity changes arising from a number of dynamical and physical possibilities. Running several versions of a global model at very high resolution, the relative contribution to the intensity drop of Lili arising from cooler sea surface temperatures, dry air advection into the storm, advective non-linear dynamics, non-advective dynamics, and shallow and deep cumulus convection was examined. This line of inquiry led to the conclusion that dry air advection from the north into the storm and the slightly cold sea surface temperatures were not the primary contribution to the observed pressure rise by 22 hPa. The primary contribution to the pressure rise was found to be the ‘rest of dynamics’ (the non-advective dynamics). The shallow convection contributed slightly to an overall cooling, i.e. a weakening of the warm core of Lili. The effects of deep cumulus convection appeared to be opposite, i.e. towards maintaining a strong storm. A primary term in the ‘rest of dynamics’, the advection of Earth’s angular momentum into the storm, is identified as a major contributor for the intensity change in the analysis. This feature resembles an intrusion of dry air into the core of the storm. This intrusion contributes to a reduction of spin and an overall rapid weakening of the hurricane. The angular momentum partitioning appears quite revealing on the sudden demise of Lili.

Abstract The 2004 eastern North Pacific hurricane season is reviewed. It was a below-average season in terms of number of systems and landfalls. There were 12 named tropical cyclones, of which 8 became hurricanes. None of the tropical storms or hurricanes made landfall, and there were no reports of deaths or damage. A description of each cyclone is provided, and track and intensity forecasts for the season are evaluated.

This paper addresses the weakening of Hurricane Lili of October 2002 just before it made landfall in Louisiana. This hurricane weakened from a category 4 storm on October 3, 2002 at 0000 UTC to a category 1 storm on October 3, 2002 at 1300 UTC. This sudden drop in intensity has been a subject of considerable interest. In this paper we explore a forecast model diagnostic approach that explores the contribution to the hurricane intensity changes arising from a number of dynamical and physical possibilities. Running several versions of a global model at very high resolution, the relative contribution to the intensity drop of Lili arising from cooler sea surface temperatures, dry air advection into the storm, advective non-linear dynamics, non-advective dynamics, and shallow and deep cumulus convection was examined. This line of inquiry led to the conclusion that dry air advection from the north into the storm and the slightly cold sea surface temperatures were not the primary contribution to the observed pressure rise by 22 hPa. The primary contribution to the pressure rise was found to be the ‘rest of dynamics’ (the non-advective dynamics). The shallow convection contributed slightly to an overall cooling, i.e. a weakening of the warm core of Lili. The effects of deep cumulus convection appeared to be opposite, i.e. towards maintaining a strong storm. A primary term in the ‘rest of dynamics’, the advection of Earth’s angular momentum into the storm, is identified as a major contributor for the intensity change in the analysis. This feature resembles an intrusion of dry air into the core of the storm. This intrusion contributes to a reduction of spin and an overall rapid weakening of the hurricane. The angular momentum partitioning appears quite revealing on the sudden demise of Lili.