SR SSD 2000-18
BRET: A MAJOR HURRICANE LANDFALL IN DEEP SOUTH TEXAS
Shawn P. Bennett
Hurricane Bret was the first major hurricane to form during the 1999 Atlantic tropical cyclone season. Bret was the only tropical cyclone to intensify to a hurricane in the Gulf of Mexico in 1999. Bret was a small hurricane that strengthened to a category 4 on the Saffir-Simpson scale (Simpson, 1974) before landfall. The lowest central pressure recorded was 944 mb on Sunday August 22 at 1500 UTC (10 am local) with sustained winds of 140 mph (122 kt). Despite its strength, Hurricane Bret was overshadowed by hurricanes Floyd and Dennis, storms notorious for their devastating affect on the lives and property of Atlantic coastal residents (ref. Figure 1). Although an intense tropical cyclone, Hurricane Bret's affects on the coastal residents of South Texas were muted by its landfall location and storm scale. In short, the disastrous effects of Hurricane Bret on the population of the region were less than expected. Disaster surveys by meteorologists from both the Brownsville and Corpus Christi National Weather Service Offices (WFOs) confirmed this assertion. This study will examine the lifecycle of Hurricane Bret and provide observational evidence for its rapid intensification and a hypothesis for its small size. Evidence and observations of Bret's strength, intensity, and damage potential at and around landfall will be presented, as well as an explanation for the limited amount of obvious physical damage wrought by the storm.
Hurricane Bret formed as a tropical depression over the Bay of Campeche on August 18, 1999 and made landfall at Padre Island National Seashore, Kenedy County, south Texas at 0000 UTC August 23, 1999. Bret was the first hurricane to make landfall on the Texas coast since Hurricane Jerry in October 1989. The last hurricane to affect south Texas was Hurricane Allen in 1980. Lawrence and Kimberlain (1999) reported that Hurricane Bret was the strongest to strike Texas since Hurricane Celia in 1970.
2.1 Lifecycle Overview: Initiation to Landfall
On August 16, a mesoscale convective complex (MCC) (Maddox, 1980) formed off the coast of Nicaragua associated with an upper-level cyclonic circulation. The MCC initiated a weak surface low over the Yucatan that combined with a tropical wave and migrated slowly toward the Bay of Campeche on August 17. This area of disturbed weather continued to intensify as an upper-level ridge strengthened over the southwest Gulf of Mexico replacing an upper-level trough providing for a more favorable weak vertical wind shear. At 2100 UTC on Wednesday August 18 the wave was classified as a tropical depression by the Tropical Prediction Center (TPC). With ship reports indicating winds in excess of 39 mph (34 kt), the depression was upgraded to Tropical Storm Bret on the afternoon of Thursday, August 19. Bret remained nearly stationary over the Bay of Campeche as an upper-level ridge remained over the southwest Gulf. During the early morning hours of Friday, August 20, Bret began a slow motion to the north. A hurricane watch was issued for the northeast Mexican coast from La Pesca to Veracruz as Bret was forecast to turn northwest during the next 24 hours. Late that evening Bret was located 215 miles east of Tampico, Mexico as maximum sustained winds increased to 74 mph (64 kt). At that time, Bret was classified as the first hurricane of the 1999 season.
On the morning of August 21, hurricane watches were extended northward to Baffin Bay, Texas as Bret continued on a more northerly track. Hurricane warnings were issued for the same area later that morning. Maximum sustained winds of Bret were near 100 mph, a category 2 on the Saffir-Simpson scale. The center of circulation was now about 300 miles southeast of Brownsville. Bret continued north through the afternoon and evening of Saturday, August 21. Consequently, hurricane warnings were extended further north to Port O' Connor, Texas. By early evening, maximum sustained winds in Bret were near 130 mph (113 kt), a strong category 3 hurricane.
The storm continued on a north-northwest track through the early morning of Sunday, August 22 with maximum winds near 140 mph (122 kt), now a category 4 hurricane. By mid-morning, the forward speed of the hurricane decreased to less than 10 mph as the eye of Bret passed just 100 miles to the east of Brownsville (ref. Figure 2). By noon, a west-northwest track had been established and landfall occurred north of Port Mansfield, Texas early that evening. Bret dissipated in the high terrain of the north central Mexican state of Coahuila. A more detailed description of Hurricane Bret can be found in Lawrence and Kimberlain (1999).
3. OBSERVATIONAL EVIDENCE
3.1 Warm sea surface temperatures
Bosart et al. (2000) and numerous authors before them have reported that hurricane intensification can be broadly related to three physical processes: 1) large-scale environmental influences, 2) storm-scale internal dynamics, and 3) ocean-atmosphere internal dynamics. In their study, Bosart et al. (2000) focused on how Hurricane Opal in October, 1995 was influenced by the interaction between the equatorward jet-entrance region of a progressive synoptic scale trough while coincidently passing over a warm-core eddy (WCE) of the loop current (ref. Figure 3). Figure 4 shows that, like Hurricane Opal, Hurricane Bret underwent a similar explosive intensification during the time that the cyclone traversed a core of offshore warm water. Conversely, Figure 4 also shows that in addition to the increased frictional effects (Burpee, 1986), near shore cool water aided in the weakening of Hurricane Bret just before and at landfall. These dramatic shifts in intensity can be attributed mainly to the ocean-atmosphere internal dynamics, which simply put, acts to charge or discharge the energy budget of a tropical cyclone. No synoptic scale trough interactions occurred during the explosive intensification period of Hurricane Bret.
3.2 Brownsville WSR-88D and NOAA P-3 Tail Doppler
Figure 5 is a reflectivity cross-section from the Brownsville WSR-88D which shows that at landfall the depth of the eyewall convection ranged between 9-10.5 km (~30,000 ft -35,000 ft). Studies by Szoke et al. (1986), on data collected during GATE (Global Atmospheric Research Program (GARP) Tropical Atlantic Experiment), showed that the typical height of the 20 dBZ reflectivity contour was about 10 km (ref. Figure 6). Dodge (1999) reported that an updraft of 13.8 ms-1 was measured by the NOAA P-3 tail Doppler radar at 0005 UTC in a spiral rainband crossing Matagorda Island. He reported that this spiral rainband had the strongest reflectivity signal observed by the NOAA P-3 tail Doppler radar during the entire research flight conducted during the landfall of Bret. Zipser and LeMone (1980) and Jorgensen et al. (1985) compared vertical velocities in updraft and downdraft cores from data collected during GATE and the Thunderstorm Project (Byers and Braham, 1949) (ref. Figure 7). These studies show that the updraft measured by Dodge (1999) was much stronger than the typical strongest 10% of updraft cores found in strong hurricanes and that it was more typical of those found in continental thunderstorms. Figure 8 shows Bret about 3 hr prior to landfall. Figure 9 shows the western eyewall of Bret over the Laguna Madre and the eye of Bret making landfall on the uninhabited Padre Island National Seashore before tracking into the sparsely populated Kenedy County. This landfall location is about 70 miles south of Corpus Christi and about 60 miles north of Brownsville. Figure 10 shows that at about 7000 ft there was a 154 kt (177 mph) storm relative velocity on the west side of the eyewall and a 128 kt (147 mph) storm relative velocity on the east side of the eyewall. Dodge (1999) reported GPS-sonde mean boundary layer (0-500 m) winds of 120 kt (138 mph) on the east eyewall. Thus, although Hurricane Bret was a small sized tropical cyclone, observational evidence suggests that it did indeed possess the destructive potential commensurate with its category 4 Saffir-Simpson scale rating.
4. DAMAGE ASSESSMENT
Shortly after Hurricane Bret made landfall, staff from NWS Corpus Christi (CRP) and NWS Brownsville (BRO) conducted separate damage surveys. Widespread flooding was observed over the Corpus Christi metropolitan area. The survey teams encountered severe flooding along Highway 77 near Driscoll. However, soils in Kenedy County tend to be sandy so that much of the initial flood water most likely infiltrated into the subsurface layer. Nevertheless, standing water could be found in previously dry washes. Wind damage was reported and observed from Port Mansfield to Kingsville to Falfurrias and was characterized as light to moderate. For example, numerous road and business signs were toppled. In Falfurrias, a few structures lost there roofs (ref. Figure 11). In Kingsville, isolated areas of roof and tree damage was noted. Across the King Ranch, which comprises most of Kenedy County, numerous windmills which pump groundwater for livestock were destroyed. Also, in Kenedy County, several high tension power lines were knocked down, as were numerous low scrubby mesquite trees, some oak trees, and a few fence lines. Near Armstrong a 310 ft radio tower was toppled. At the INS Inspection Station near Sarita sheet metal was found twisted and sheared off of the overhead awning. In Port Mansfield, damage was observed in a trailer park where some trailers had been overturned and other structures had roof damage. However, most of the structures remained intact. Some Port Mansfield residents seemed genuinely surprised by the lack of damage, especially given that Bret had been classified a category 4 hurricane. Further south along US Highway 77 toward Raymondville relatively little significant damage was observed other than downed tree limbs.
5.1 Bret Size and Intensity Limitations
A qualitative review of tropical cyclones that have affected Texas over the past 100 years showed that those cyclones that originate over the tropical Atlantic or Caribbean are generally the stronger, while those that originate over the Bay of Campeche or the margins of the Gulf of Mexico are generally the weaker. We hypothesize that given a barotropic atmosphere and a lack of warm plumes of water or WCE, that the inherent geographic restriction of the Bay of Campeche and margins of the Gulf of Mexico often limit the scale and intensity of tropical cyclones that originate there. Whereas tropical cyclones that originate in the tropical Atlantic are usually not limited by space and time so that they often reach their maximum potential intensity (Emanuel, 1986). Exceptions to this hypothesis include storms like Bret or Opal that encounter oceanic and or atmospheric influences that interact to rapidly energize the cyclone. Thus, these influences provide a mechanism to overcome the natural space and time scale limitations of their geographic origin. Figure 12 shows visible satellite imagery of Hurricane Bret at the time of its maximum intensity at 1433 UTC on August 22. The southern extent of the central dense overcast surrounding the eye to the northern edge of the outer spiral rainband the cyclone measures about 300 miles across. By comparison, Hurricane Floyd was at its maximum intensity more than double the size of Bret. In fact, Hurricane Floyd was nearly three times the size of Hurricane Andrew. Hurricane Floyd originated in the tropical Atlantic Ocean in early September, 1999 and Hurricane Andrew originated in the tropical Atlantic in mid-August, 1992.
5.2 Comparison to Recent South Texas Hurricanes
Although several hurricanes have affected the south Texas coast during the past 50 years, the origin and track of Bret were the unique features of this hurricane. Bret was the only hurricane in recent history to completely develop over the Bay of Campeche. Hurricane Alice developed over the southwest Gulf of Mexico in June 1954, just north of the Bay of Campeche. Alice made landfall just south of Brownsville and produced heavy rains over the Rio Grande River Valley and Plains. Hurricane Celia formed in late July 1970 over the western Caribbean then tracked to the northwest into the central Gulf of Mexico where it gained hurricane strength before making landfall near Corpus Christi Texas. Other intense hurricanes, such as Beulah (1967), Allen (1980) and Gilbert (1988) reached hurricane strength over the western Atlantic before entering the Gulf of Mexico. Bret, like most hurricanes, produced widespread torrential rains along its path. Radar-estimated storm total precipitation over parts of northern Kenedy ranged from 15 to 20 inches (ref. Figure 13). Widespread flooding was reported in the Corpus Christi metropolitan area as Bret moved inland on the evening of Sunday, August 22. Although flooding from Bret was significant, several hurricanes produced more catastrophic flooding. For example, widespread rainfall with Beulah ranged between 25 and 30 inches over a large part of south Texas. This resulted in record river flooding along the Lavaca, Navidad, San Antonio and Rio Grande rivers.
Although Bret intensified to a category 4 hurricane, wind damage and higher storm surge were more prevalent with several other lower category storms that struck south Texas. The best example is Hurricane Allen which was rated as a category 3 cyclone and produced sustained winds well in excess of 100 mph (87 kt) along much of the south Texas Coast. With Allen, storm surge reached 12 ft in the Port Mansfield area. Yet, despite the passage of the eye of Bret just north of Port Mansfield, no significant widespread storm surge nor beach erosion occurred. The only significant beach erosion and storm surge was isolated to very near where the eye made landfall on Padre Island. In this limited area, new cuts in Padre Island were observed by LANDSAT an aerial survey along and near where landfall occurred (ref. Figs. 14 and 15).
We have shown that Hurricane Bret was a powerful cyclone that had tremendous capacity to do damage. Figures 14 and 15 demonstrate the devastating impact of the landfall of Bret in the form of newly created cuts across the island. We have shown that because of its compact size, its passage across the near shore cool waters, and landfall on the uninhabited Padre Island National Seashore combined with its path through the sparsely populated Kenedy County, about halfway between Brownsville and Corpus Christi, the impact on the major population centers was mitigated. In summary, the majority of South Texans were quite fortunate that Hurricane Bret was a small cyclone and that it followed the track that it did.
The authors wish to thank Hector Guerrero and Richard Hagan for their careful review of the manuscript.
Bosart, L.F., C.S. Velden, W.E. Bracken, J. Molinari and P.G. Black, 2000: Environmental influences on the rapid intensification of Hurricane Opal (1995) over the Gulf of Mexico, Mon. Wea. Rev., 128, 322-352.
Byers, H.R., and R.R. Braham, Jr., 1949: The Thunderstorm. Supt. of Documents, U.S. Government Printing Office, Washington, D.C. 287 pp.
Burpee, R.W., 1986: Mesoscale structure of hurricanes. Mesoscale Meteorology and Forecasting, P.S. Ray, Ed., Amer. Meteor. Soc., 311-330.
Dodge, P., 1999: Mission Summary: Hurricane Bret Landfall 9908022I Aircraft: 43RF, NOAA/OAR/HRD, mission summary report, 10 pp.
Emanuel, K.A., 1986: An air-sea interaction theory for tropical cyclones. Part I: Steady-state maintenance. J. Atmos. Sci., 43, 585-604.
Jorgensen, D.P., E.J. Zipser, and M.A. LeMone, 1985: Vertical motions in intense hurricanes. J. Atmos. Sci., 42, 839-856.
Lawrence, M.B., and T.B. Kimberlain, 1999: Preliminary Report: Hurricane Bret 18-25 August 1999, NOAA Tropical Prediction Center, National Hurricane Center, pp.10.
Maddox, R.A., 1980: Mesoscale convective complexes. Bull. Amer. Meteor. Soc., 61, 1374-1387.
Simpson, R.H., 1974: The hurricane disaster potential scale. Weatherwise, 27, 169-186.
Szoke, E.J., E.J. Zipser and D.P. Jorgensen, 1986: A radar study of convective cells in GATE. Part I: Vertical profile statistics and comparison with hurricane cells. J. Atmos. Sci., 43, 184-197.
Zipser, E.J., and M.A. LeMone, 1980: Cumulonimbus vertical velocity events in GATE, Part II: Synthesis and model core structure. J. Atmos. Sci., 37, 2458-2469.
1. Corresponding author address: Shawn P. Bennett, National Weather Service, 20 South Vermillion Road, Brownsville, Texas 78521-5798; e-mail: firstname.lastname@example.org