[From the U.S. Government Printing Office, www.gpo.gov]
Service Assessment f- AA '4@ A- P tip 'Vo July 5 -141) 1996 U.S. Department of Commerce 02 NrMosfj National Oceanic and Atmospheric Administration I Weather Service I -I,URRI , NE-:d BE ,pring, MD C. QC 945 )97 NT Of- .u5 H87 1997 i Cover photograph: Hurricane Bertha - http.,Ilwww.ncdc.noaa.govlpubldatalimageslI Service Assessment URRICAN July 5 -14, 1996 Property of CSC Library P"FMOSPI" U.S. Department of Commerce National Oceanic and Atmospheric Administration National Weather Service Silver Spring, MD C, BERTHA April 1997 US Department of commerce NENT Of- NOAA Coastal Services center Library 2234 South Hobson Avenue Charleston, SC 29405-2413 Preface The primary purpose of this Service Assessment is to document the evaluation of the National Weather Service's (NWS) performance in fulfilling its mission of providing timely warnings and accurate forecasts for Hurricane Bertha. The NWS's products and services used by emergency managers are key to preparedness for and the mitigation of a tropical cyclone's impact. More specifically, this was the first time a complete Hurricane Liaison Team (HLT) was used in an operational environment. This team is ajoint venture between the Federal Emergency Management Agency (FEMA), the NWS and state and local emergency management officials. The concept is to supplement the National Hurricane Center (NHC) staff with Federal, state and local emergency management personnel to provide information and coordination between the NHC and emergency management decision makers. As this assessment has noted, the warning process is a partnership between the NWS and all organizations charged with responding to natural hazards. We in the NWS will continue to forge and nurture relationships to ensure the best possible warning service for our citizens. Elbert W. Friday, Jr. April 1997 Table of Contents Page Preface .................................................................... ii Acronyms .................................................................. iv Acknowledgments ........................................................... vi Executive Summary .......................................................... ix Summary of Findings and Recommendations ....................................... xii Chapter 1 Service Assessment Summary ..................................... I Chapter 2 National Perspectives ............................................. 5 National Hurricane Center ................................... 5 Hydrometeorological Prediction Center ........................ 11 Marine Prediction Center .................................... 12 Storm Prediction Center ..................................... 16 Chapter 3 Regional Perspectives ............................................ 19 Southern Region ........................................... 19 Eastern Region ............................................ 23 Appendices Appendix A Saffir-Simpson Scale ............................................ A-1 Appendix B Meteorological Tables ........................................... B-I Appendix C Fujita Scale .................................................... C-1 Appendix D Tornado Watch Areas ........................................... D-I Appendix E Aerial Survey .................................................. E-I iii Acronyms AFOS Automation of Field Operations and Services AOC Aircraft Operations Center ASOS Automated Surface Observing System AST Atlantic Standard Time AVN Aviation Model CWA County Warning Area DEMA Delaware Emergency Management Agency ECMWF European Center for Medium Range Weather Forecasting EDT Eastern Daylight Time EMC Emergency Management Center EMWIN Emergency Managers Weather Information Network ERH Eastern Region Headquarters ET Electronics Technician Eta Eta Model FAA Federal Aviation Administration FEMA Federal Emergency Management Agency ft feet GFDL Geophysical Fluids Dynamical Lab Model GOES Geostationary Operational Environmental Satellite HLS Hurricane Local Statement HLT Hurricane Liaison Team HPC Hydrometeorological Prediction Center in inch IR Infrared kts knots MAR' Modernization and Restructuring mb, millibar MIC Meteorologist in Charge MPC Marine Prediction Center mph miles per hour MRF Medium-range Forecast Model MRFX Medium-range Forecast Experimental Model NAWAS National Warning System NCEP National Centers for Environmental Prediction NEXRAD Next Generation Weather Radar NGM Nested Grid Model NHC National Hurricane Center nm nautical miles NOAA National Oceanic and Atmospheric Administration NOGAPS Naval Operational Global Atmospheric Prediction System iv NOW Short Term Forecast NWR NOAA Weather Radio NWS National Weather Service NWSFO NEXRAD Weather Service Forecast Office NWSO NEXRAD Weather Service Office NWWS NOAA Weather Wire Service OM Office of Meteorology PUP Principal User Processor QPF Quantitative Precipitation Forecast RDA Radar Data Acquisition Soo Science and Operations Officer SPC Storm Prediction Center SRH Southern Region Headquarters SSMI Special Sensor Microwave Imagery TPC Tropical Prediction Center TWC The Weather Channel UTC Coordinated Universal Time VIS Visible VITEMA Virgin Islands Territory Emergency Management Agency WCM Warning Coordination Meteorologist WFO Weather Forecast Office WSH Weather Service Headquarters WSR-88D Weather Surveillance Radar-1988 Doppler WV Water vapor v Acknowledgments This report was produced by having those responsible for each portion of the service provided write their own section. Each NWS field office and NWS Regional Office affected by Bertha conducted its own survey of the services they provided. Additionally, assessments were written by the National Centers that provided products and services related to Bertha. The Service Assessment Team is grateful to all those that provided valuable input into this document. The individuals primarily responsible for the Assessment are as follows: Service Assessment Team Rick Watling, Products and Services Manager, Meteorological Services Division, NWS Eastern Region Headquarters (ERH), Bohemia, New York Gene Hafele, Warning Coordination Meteorologist (WCM), NEXRAD (Next Generation 'Weather Radar) Weather Service Office (NWSO), Houston, Texas ,Scott Kiser, Technical Leader, Customer Service, Office of Meteorology (OM), Weather Service Headquarters (WSH), Silver Spring, Maryland NWS National Centers Robert Burpee, Director, Tropical Predication Center (TPQ, Miami, Florida Miles Lawrence, Hurricane Specialist, NHC, Miami, Florida Mike Schichtel, Meteorologist, Hydrometeorological Prediction Center (HPC), Camp Springs, Maryland Dave Feit, Chief Marine Forecast Branch, Marine Prediction Center (MPC), Camp Springs, Maryland Fred Ostby, Chief Operational Guidance Branch, Storm Prediction Center (SPQ, Kansas City, Missouri vi Weather Service Headquarters Rainer Dombrowsky, Customer Service, OM, WSH, Silver Spring, Maryland Linda Kremkau, Technical Editor, Customer Service, OM, WSH, Silver Spring, Maryland NWS Regions and Field Offices Gary Woodall, WCM, Southern Region Headquarters (SRH), Fort Worth, Texas Harvey Thurm, Synoptic Scale Meteorologist, ERH, Bohemia, New York Brian Smith, WCM, NEXRAD Weather Service Forecast Office (NWSFO), Omaha, Nebraska Steve Rich, Meteorologist in Charge (MIC), NWSO, Charleston, South Carolina Jerry Harrison, WCM, NWSO, Charleston, South Carolina Bernie Palmer, MIC, NWSFO, Columbia, South Carolina Steve Naglic, WCM, NWSFO, Columbia, South Carolina Richard Anthony, MIC, NWSO, Wilmington, North Carolina Tom Matheson, WCM, NWSO, Wilmington, North Carolina Tom Krielm, MIC, NWSO, Morehead City, North Carolina Dan Bartholf, WCM, NWSO, Morehead City, North Carolina Steve Harned, MIC, NWSFO, Raleigh, North Carolina George Lemons, WCM, NWSFO, Raleigh, North Carolina Tony Siebers, MIC, NWSO, Wakefield, Virginia Hugh Cobb, Science and Operations Officer (SOO), NWSO, Wakefield, Virginia Bill Sammler, WCM, NWSO, Wakefield, Virginia Jim Travers, MIC, NWSFO, Sterling, Virginia vii Gary Szatkowski, Deputy MIC, NWSFO, Sterling, Virginia Barbara Watson, WCM, NWSFO, Sterling, Virginia Chet Henricksen, MIC, NWSFO, Philadelphia, Pennsylvania Joe Miketta, WCM, NWSFO, Philadelphia, Pennsylvania Mike Wyllie, MIC, NWSFO, New York City, New York Gary Conte, WCM, NWSFO, New York City, New York Bob Thompson, MIC, NWSFO, Boston, Massachusetts Dave Vallee, Service Hydrologist, NWSFO, Boston, Massachusetts Glenn Field, WCM, NWSFO, Boston, Massachusetts Al Wheeler, MIC, NWSFO, Portland, Maine John Jensenius, WCM, NWSFO, Portland, Maine Patricia Haas, Hydrologist, Mid Atlantic River Forecast Center, State College, Pennsylvania Israel Matos, MIC, NWSFO, San Juan, Puerto Rico Rafael Mojica, WCM, NWSFO, San Juan, Puerto Rico Paul Hebert, MIC, NWSFO, Miami, Florida Jim Lushine, WCM, NWSFO, Miami, Florida Bart Hagemeyer, MIC, NWSO, Melbourne, Florida Dennis Decker, WCM, NWSO, Melbourne, Florida Steve Letro, MIC, NWSO, Jacksonville, Florida Fred Johnson, WCM, NWSO, Jacksonville, Florida viii Executive Summary Background Hurricane Bertha was an early season Category 2 storm when it made landfall on the coast of North Carolina on July 12, 1996. Twelve deaths have been directly attributed to Bertha with United States damages estimated at $270 million. Extensive evacuations of vulnerable areas occurred in advance of Bertha, including 250 million in North Carolina, 80,000 in South Carolina and 20,000 in Georgia. Revenue losses to the tourist industry approached $40 million. Bertha closely paralleled the southeast U.S. coast at a distance from 170 to 200 miles. This created multiple state involvement with almost all of the U.S. east coast involved with some watch or warning. The NHC's track forecast was very accurate with average forecast track errors 15 percent lower than the 10-year official track averages. Where the storm came ashore in North Carolina, watches and warnings were posted 65 hours and 47 hours before landfall, respectively, which far exceeds the NHC stated goals of 36 and 24 hours. Nevertheless, the coordination between NHC and emergency managers, regarding the issuances of watches and warnings, was frustrated by efforts to reconcile the meteorology of the event with state and local response requirements. Bertha's anticipated turn from a northwest direction to north-northwest as it approached the southeast coast was agonizingly slow. This put a great deal of pressure on the elected officials and Emergency Management Centers (EMCs) of Florida, South Carolina and North Carolina concerning what actions to take. This was also the first time the HLT had been fully deployed at the NHC. A Service Assessment Team was dispatched to the NHC as well as the impacted states to investigate these issues and to evaluate the HLT. Issues Bertha's slow northward turn off the Florida coast caused the state of Florida to urge NHC to issue watches when NHC felt they were unnecessary. The major concern was that if the storm continued to move differently than forecast, sufficient response time would not be available if watches or warnings were issued at short notice. Similarly, South Carolina expressed the need for early release of the official forecast track before NHC had completed the internal NWS coordination process. Several recommendations have been made in this report to help resolve theseissues. A variety of communication/coordination issues were also revealed. NHC/state briefings were complicated by the fact that no single communication system was used to brief state emergency management officials, adding to the workload of the hurricane specialists and confusion between state EMCs and the NHC as to what method of coordination would be employed. Similar ix coordination problems occurred between NWS field offices and local EMCs as well as between some state and local EMC(s). Internal coordination between NWS offices posed difficulties as not all necessary offices are connected to the Hurricane Hotline. This problem will continue to increase as additional NWS offices spin up to full Weather Forecast Office (WFO) status. The amount of meteorological information received by state and local EMCs to a large extent depended upon the resources available to each office. Even though it is not considered an operational system, Internet received high praise while the National Oceanic and Atmospheric Administration (NOAA) Weather Wire Service (NWWS) continues to prove too costly for some EMCs. Additionally, NOAA Weather Radio (NWR) coverage is still lacking in some coastal areas. Accordingly, the NWS should seek to expand implementation of the Emergency Management Weather Information Network (EMWIN) as well as NWR. This was the first time the HLT was used in an operational environment. This team is ajoint venture between FEMA, the NWS and state and local emergency management officials. The concept is to supplement the NHC staff with Federal, state and local emergency management personnel to provide information and coordination between the NHC and emergency management decision makers. As this is a new initiative, the HLT is still evolving in response to the needs and desires of both the partners in the HLT and their customers. Questions remain regarding the ultimate mission, function and operations of the HLT. Additionally, a larger resource pool of available people to staff the HLT needs to be identified, and smoother procedures for setting up the team's hardware need to be established. Some data acquisition problems were also noted. The St. Croix and St. Thomas Automated Surface Observing Systems (ASOS) were both powered down by the Federal Aviation Administration (FAA) when the airport towers were abandoned resulting in the loss of valuable ground truth data. Due to the remoteness of the site, the San Juan Weather Surveillance Radar- 1988 Doppler (WSR-88D) remote data acquisition (RDA) unit uses generator power since commercial power is not available. The generator produces irregular frequency fluctuations which puts the system out of service. The FAA electronics technician was able to keep the system operational throughout the storm, and a more permanent fix was employed after the .passage of Bertha. NWSFO Miami noted that without data off the southeast Florida coast, no wave information is available south of Melbourne. This forced the staff to estimate waves from their wind forecasts. Meteorology Bertha was -an early season Cape Verde hurricane which started as a tropical depression in the central tropical Atlantic on July 5. For 3 days, the depression moved west-northwest at 23 to 29 miles per hour (mph) and strengthened to a hurricane on July 8 as the center moved across the Leeward and Virgin Islands of the northeast Caribbean. Bertha turned northwestward on July 9. Maximum sustained winds reached 115 mph at 2 a.m. Eastern Daylight Time (EDT), when x Bertha was centered 138 miles north of Puerto Rico. The last hurricane to reach this strength this early in the season was Alma in 1966. Moving northwest, the center paralleled the Bahama islands. The track became north-northwest on July 10 and I I as the center moved parallel to the coast of Florida and Georgia. With a forward speed of 17 mph, Bertha made landfall at 4 p.m. EDT on July 12 on the coast of North Carolina, midway between Wrightsville and Topsail Beaches. Bertha quickly dropped below hurricane strength when it moved inland over eastern North Carolina. Highest wind gusts reported on land were 108 mph at Camp Lejeune and Jacksonville, North Carolina. The lowest observed sea-level pressure at landfall was 977 millibars (mb) at Surf City, North Carolina. A value of 974 mb is assumed to be the minimum landfall pressure. Bertha then moved northeast along the U.S. east coast, producing 45 to 60 mph sustained winds over land from northern North Carolina to New England. Bertha was declared extratropical on July 14 when the center moved from the Maine coast to New Brunswick, Canada. Coastal storm surge flood heights, from Florida through New England, ranged from I to 4 feet, but values to 5 feet were estimated on the North Carolina coast from Cape Fear to Cape Lookout. A storm surge of 6 feet was indicated near Swansboro, North Carolina. Rainfall associated with Bertha occurred from the eastern Carolinas to Maine with many locations receiving 2- to 4-inch amounts. Isolated reports of over 6 inches were reported from North Carolina to Massachusetts. The maximum was 7.20 inches at Billerica, Massachusetts. Bertha's rapid motion through the Northeast served to limit rainfall duration and amounts. Six tornadoes have been confirmed with four in Virginia, one in North Carolina and one in Maryland. xi Summary of Findings and Recommendations National Implementation Finding 1a: Some confusion remains as to the purpose of the HLT. Finding 1b: The workload for the hurricane specialists, with the HLT in place, was not reduced and in some instances increased. Recommendation 1: In conjunction with FEMA, the functions and operations of the HLT must be reviewed and refined. The results should be briefed at the National Hurricane Conference and other suitable forums. FEMA representatives are encouraged to take an active role in the Interdepartmental Hurricane Conferences. Finding 2: There are insufficient HLT members to support speciality functions (electronics, communications) to avoid burnout, turnover and absences due to sickness, annual leave, etc. Recommendation 2: As part of the NWS/FEMA review, the specific tasks of the HLT members must be outlined and an adequate resource pool identified. The NOAA/NWS Hurricane Conference should be the forum for discussing and validating HLT mission, membership and other requirements. Finding 3: Hardware required to support the HLT is extensive and requires significant time and effort to set up. Recommendation 3: FEMA must provide a complete set of instructions that will assist in the set-up process. Additionally, FEMA and NHC should identify electronic specialist(s) who can set up and test the equipment prior to the arrival of the HLT. Given the location of the NHC on the Florida International University campus, NHC could explore the possibility of involving university resources to assist in the process. xii Finding 4a: There was no single communication system used to brief state emergency management officials. This led to confusion among the EMCs. Finding 4b- NHC was required to conduct multiple briefings when more than one state was impacted by Hurricane Bertha. This proved to be extremely time consuming for the hurricane specialists. Finding 4c: During teleconferences established by the HLT, the local NWS offices were not always involved. Finding 4d: Several times local emergency managers were tied into state and regional briefings coordinated through the HLT. The locals would then be brought into a local conference call with their state and local NWS offices. This was repetitive and not a good use of the local emergency managers limited time. Finding 4e: NHC briefing calls to North Carolina state emergency management officials frequently came at times when these individuals were already in the midst of a briefing by the NWSFO or with local emergency management officials along the coast. Recommendation 4: FEMA and the NWS must ensure that coordination methodologies and protocols are firmly established before the next hurricane season. The local NWS office(s) must be included in all NHC state conference calls to add local expertise and to prepare for their coordination calls with local emergency managers. Finding 5a: The Modernization and Restructuring (MAR) of the NWS has created a greater need for internal forecast coordination as the number of forecast offices has increased. Finding 5b: As Hurricane Bertha approached the Delmarva region,confusion arose when the wind forecasts from two NWS offices for adjacent forecast areas were significantly different. Recommendation 5: The NWS needs to move quickly to establish an internal coordination mechanism. xiii Finding 6: Some EMCs are having problems receiving weather information. NWR coverage is incomplete and NWWS is too costly for some EMCs. Computer bulletin boards and the Internet received high praise. Recommendation 6: The NWS should continue to seek partnerships to expand NWR coverage in support of the Gore initiative and target resources to ensure reliable programming and maintenance of the new systems. The &WS must actively support implementation of EMWIN and seek additional partnerships to ensure its availability from multiple sources. EMCs should be reminded of other resources, such as the free redistribution of NWWS from their state distribution point and the use of Packet Radio. Finding 7a: A number of offices noted a truncation of some of the more lengthy Hurricane Local Statements (HLS) on The Weather Channel (TWC). Moreover, it was also noted that some of the Short Term Forecasts (NOW) were cut off at six, as opposed to eight lines. This premature truncation of the HLS and NOW resulted in critical information not being passed on to the vie w'er. Finding 7b The "segmented" HLS format used by NWSFO San Juan marked an improvement over their HLSs issued during Hurricane Marilyn. Finding 7c: NWSFO San Juan's NOWs were informative, but some of them tended toward the radar narrative summary style of writing. Recommendation 7: OM needs to work with TWC to resolve the truncation issue. All offices should be reminded of the NOW's primary function, to serve as a forecast of future conditions and impacts rather than a summary of past events. Finding 8: Potentially valuable meteorological data was lost when the St. Croix and St. Thomas ASOS units were powered down when the FAA abandoned their towers. Recommendation 8: NWS and FAA officials should ensure that ASOS units remain in service during the approach of a tropical cyclone. xiv Finding 9: Rip currents associated with Bertha resulted in 3 deaths and over 100 rescues along the Florida and southeast Georgia coasts. Heavy surf advisories were canceled before two of the deaths occurred. Lack of data off the southeast Florida coast forced forecasters to estimate wave heights from local wind forecasts. Recommendation 9: NWSFO Miami must redouble their efforts at seeking partnerships with Beach Patrol units and other organizations to secure critical wave height and surf reports. Finding 10: Hurricane specialists were urged to change the location of watches and warnings. This was exemplified when Bertha's slow northward turn resulted in Florida State officials requesting NHC to post watches on the Florida coast when hurricane specialists felt that this was unnecessary. Recommendation 10: NHC needs to remain sensitive to unique and special situations that occur which emergency managers must resolve during hurricane threats. State officials have the authority to issue mandatory evacuations for vulnerable coastal areas even when NHC may feel that meteorological conditions do not justify issuing hurricane watches or warnings. The NWS and FEMA need to expand the present hurricane course for emergency managers and should develop a distance learning approach to the present NHC resident course. As a part of this course, a module should be developed for NWS personnel at both the NHC and at local offices to ensure that they fully appreciate emergency manager requirements. Finding 11: The NHC was requested to release forecast points early. Recommendation 11: NHC should not release forecast points prior to completion of forecast coordination. If the meteorology changes significantly and unexpectedly between the 6 hourly advisory forecasts, then NHC will issue a special advisory. However, in truly unique circumstances, some coordination on trends and changes from the previous forecast can be discussed with the local NWSO and NWSFO so they can brief state or local EMCs prior to the issuance of the new forecast. There is also a continuing need to educate emergency managers and other customers about hurricane forecasts and products and the uncertainties related to these products. xv Finding 12: Hurricane strike probabilities are still not fully understood by state and local EMCs and are being misrepresented by the media. These probabilities are especially confusing when a hurricane is paralleling the coast. Recommendation 12: The NWS and FEMA should develop new educational materials related to the use of hurricane strike probabilities as well as the use of forecasts and forecast uncertainties in the decision-making process. This should be included in the residence training at NHC as well as in distance learning modules for emergency managers. These materials should also be provided to coastal WCMs for outreach efforts to the local media. Similarly, NHC should investigate better ways of graphically depicting forecasts and forecast uncertainties. Finding 13: Many NWS offices par ticipate in post-storm local-level evaluations and critiques, usually led by the WCM and/or the MIC. Recommendation 13: This practice of local self-evaluation should be encouraged at all NWS offices, coastal or otherwise. NCEP Implementation (National Centers for Environmental Prediction) Finding 14: The Internet has become very popular, and the information available over the Internet from NHC received very favorable comments. Naturally, when the hurricane was close to land, access to the NHC Homepage was difficult. Recommendation 14: NHC should investigate the possibility of increasing their Internet capabilities. Finding 15: The NHC rarely sought input from the SPC nor made attempts to ensure that the SPC was "on the line" during the hotline coordination. As a result, field offices that were concerned with the tornado threat were never sure that the SPC was on the line although the SPC makes it a practice to monitor the line continuously. Recommendation 15: The NHC should include the SPC in its roll call list for coordinating advisories when a tropical cyclone threatens the U.S. mainland. In addition, the NHC should make it a point to invite comments from the SPC in such situations. xvi Regional Implementation Finding 16a: Due to the remoteness of the location, the San Juan WSR-88D RDA does not utilize commercial power. Finding 16b: The RDA's emergency generator produces an electric current with irregular frequency fluctuations. Recommendation 16: Officials from the SRH and Puerto Rico emergency management should encourage the FAA to develop a plan for supplying the WSR-88D with stable commercial power and with stable, reliable emergency generator power. Finding 17: The lack of reliable redissernination systems caused some delay in the receipt of weather products necessary to the emergency management decision-making process in Delaware. The use of the fax blast was not always reliable. Recommendation 17: ERH should work with FEMA Region 3 and WSH to resolve this issue. xvii Chapter 1 Service Assessment Summary Operations of the Hurricane Liaison Team Traditionally, local emergency managers from Florida, as well as some FEMA regional personnel, volunteered to work at NHC to field calls from local emergency managers and to keep FEMA national and regional offices apprised of the evolving situation from an emergency management perspective. The 1996 hurricane season was the first real effort to formalize the ad hoc process and make it more of a functioning unit. This resulted from the highly active 1995 hurricane season where it became evident that something had to be done to take some of the external pressure off of the NHC hurricane specialists. Accordingly, the HLT was formed at the request of the State of Florida, FEMA and the NWS. The HLT was first activated during Hurricane Bertha. Since this was the first official activation, some of its purpose and function were not clearly understood by all of the parties involved. It was expected that the HLT could relieve some of the external demands placed on the hurricane specialist. During Bertha, this was not the case. In fact, the workload was even greater due to the number of briefings that required participation of a specialist. The HLT concept is still under development and its functions and operations are being reviewed. A major concern is to ensure that the HLT does not detract from local emergency managers contacts with their local NWS offices and WCMs. Prior to the start of the 1996 hurricane season, individuals were identified and chosen who had the necessary skills and expertise. After working Hurricane Bertha, the team realized that, given the extended hours that are required and the possibility of a team member(s) being unavailable, a sufficient number of skilled people were not identified to fill these positions. In addition to staffing, a unique set of hardware is required to meet the operational requirements of the HLT. Since the equipment was not prepositioned and there was no dedicated electronics expert assigned to assemble the hardware, some time was lost setting up the equipment. The video conferencing between NHC and FEMA Headquarters ensured that FEMA senior management and all organizations supporting the Federal Response Plan had the necessary information to be proactive, making timely response actions. Video conferences were scheduled twice a day for the Director of FEMA and were accomplished with minimal problems. State and Local Coordination Florida, South Carolina, North Carolina -- Unlike FEMA Headquarters and NHC, most state EMCs currently do not have video conference capability but expressed a willingness to add this technology. However, due to high costs, video conferencing is not affordable to NWS field offices. Most briefings are done via voice only with each state using differing communication conferencing systems to speak to the NHC. Some use the National Warning System (NAWAS); others, such as Florida, have their own conference systems with a line to NHC. Some of the conference calls were done on NAWAS while others were made through existing Federal and state teleconference systems but established by government operators. The South Carolina EMC missed an NHC briefing since they were not sure whether the call was on NAWAS or some other system. NHC and the states agreed that a universal conferencing system needs to be established as coordination is too difficult with multiple systems. A single system would also help decrease the workload for the hurricane specialist as all state EMCs could be conferenced together and briefed at once. Local NWS offices stated they were not always tied into conference calls between the state EMCs and the NHC. During Hurricane Bertha, many of the local EMCs were tied into HLT briefings along with their state EMCs. After this briefing, they would then participate in another conference call, involving EMCs in the local area, the state EMC and the local NWS. In survey team discussions with local emergency managers, they expressed little interest in talking directly with NHC. They would rather be tied into a telephone conference call with their surrounding EMCs and local NWS office. They assume the state and local NWS offices will talk with the NHC and pass information to them during their local conference call. In interviewing local emergency managers, the survey team was told NWWS continues to prove too expensive for local EMCs with limited funding. The EMWIN was discussed and could prove to be a viable alternative for users who currently have no level of support or can afford very little. North Carolina officials note that the NWR is still not available in some areas along their coastline. The use of Internet is increasing and South Carolina and North Carolina are using the Internet as one method to access the latest information. However, when Bertha approached the coastline, the Internet connection to the NHC was difficult or impossible to get. Delaware and Maryland -- NAWAS was used as the primary communication and coordination tool by Mt. Holly NWSFO during events of Hurricane Bertha. Delaware Emergency Management Agency (DEMA) felt the support provided during Hurricane Bertha did not meet the level of support provided prior to the policy change. The use of a single conference call by several NWS offices for coordination among multiple states, as well as the loss of access to the 2 26 local municipalities on the DEMA bridge, was a major concern. Subsequent work between the OM, ERH and DEMA has found an equitable resolution using the DEMA bridge. Delaware does not have an effective mechanism in place for the redistribution of both national and local weather information and guidance. This put county and municipal emergency managers at a disadvantage as they had no hard copies of NWS watches, warnings and statements. This forced coordination calls to be longer as they became a dissemination mechanism rather than a coordination tool. Delaware officials also noted several occasions where wind forecasts issued for adjacent areas by different NWSFOs were significantly different which caused confusion and raised the anxiety of many local emergency managers. NWS Operations As Hurricane Bertha paralleled the coast, it had the potential to make its initial landfall anywhere from Florida to North Carolina. This put extreme pressure on NHC as well as the state EMCs. Although NHC's forecast tracks for Bertha proved to be quite accurate, state EMCs urged NHC to change the location and timing of watches and warnings and to release the forecast tracks early. This caused tension between NHC and their external users. The state EMCs.thought their input and concerns were being ignored while NHC felt that they were being second guessed on their meteorology. While the states agreed NHC is the technical leader and the recognized expert, the EMCs continue to emphasize that watches and warnings are not purely a meteorological call but must also consider emergency response and public safety. The states agree they have the authority to issue mandatory evacuations for vulnerable coastal areas. They said that despite their efforts to motivate people, the public will often not respond to recommendations or evacuation information unless there is a watch or warning in effect. Graphical representation of hurricane strike probabilities can be a useful tool, however, it was noticed during Bertha that these graphics were not fully understood by the public or the EMCs. In some cases, it lead to misinformation, and in others the media was incorrectly depicting strike probabilities. While strike probabilities work well with storms that strike the coast perpendicularly, they were confusing when a storm parallels the coast. 3 Chapter 2 National Perspectives National Hurricane Center Synoptic History Hurricane Bertha originated from a tropical wave which moved from Africa to the Atlantic on July 1, 1996. A weak circulation was first detected on satellite imagery on July 3, centered about 575 miles south of the Cape Verde Islands in the far eastern Atlantic Ocean. The track of the circulation center begins on July 5, when the circulation is believed to have reached the surface and become a tropical depression in the central tropical Atlantic. This track is displayed in Figure I and listed in Appendix B, Table 1. Bertha followed a fairly smooth curved path around the western periphery of the Atlantic subtropical high pressure ridge. This ridge changed little during Bertha's existence, and a weak mid-level trough persisted in the western North Atlantic. For three days, the depression moved toward the west-northwest at the fast forward speed of 23 to 29 mph. It strengthened to a hurricane with 1-minute maximum sustained winds of 86 mph on July 8 as the center moved across the Leeward and Virgin Islands of the northeastern Caribbean. The center moved between Antigua and Barbuda at 2 a.m. Atlantic Standard Time. (AST) (0600 Coordinated Universal Time [UTCI) on July 8, across St. Barthelemy, Anguilla and St. Martin, just north of St. Thomas, and over the British Virgin Islands by 2 p.m. AST (1800 UTC). The track gradually turned northwestward on July 9 as maximum sustained winds reached 115 mph at 2 a.m. AST (0600 UTC). Bertha was centered 138 miles north of Puerto Rico at this time but earlier passed within 35 miles of the island. Since the strongest winds were located in the northeast quadrant of the hurricane, most of Puerto Rico experienced only tropical storm conditions. The exception is the Puerto Rican island of Culebra, where hurricane-force winds are believed to have occurred. Moving northwestward at a slower forward speed of 17 to 23 mph, the center of Bertha moved parallel to the Bahama Islands, passing 45 to 70 miles northeast of the Turks and Caicos Islands, San Salvador, Eleuthera and the Abacos. Again, the strongest winds were located to the northeast of the center, but 75-mph sustained winds might have reached some of the above mentioned islands. Continuing on its gradual turn, the track became north-northwestward on July 10 and I I as the center moved parallel to the coast of Florida and Georgia at a distance of 170 to 200 miles 5 t T I . . . . I-C VERIM .10V Hurrtcano Troptcal Storn ..... Troptcal D*pr. 55 + + Extratropical A A Subtroptcal Stn .... ..... A A Subtropical Dep OPosittoin D 12 UTC 50" : Psn/Date 6 06 UIC .. .... .. . PPP Mn. Pros. Cnb)l 10 1-) tXV 77)Uldr-l" T.- r Pr .... .... .... ... Iz- 40 00 f . ... .... .... . .... .... dd c 44, .... ...... i iwi Lee 95 go 85 as 75 70 65 66 55 56 45 46 35 36 offshore. During this time, the forward speed slowed to about 9 mph. Moving northward and re- .accelerating to a forward speed of 17 mph, Bertha made landfall at 4 p.m. EDT (2000 UTC) on July 12 on the coast of North Carolina. At landfall, Bertha was a Category 2 storrn on the Saffir- Simpson Hurricane Scale (see Appendix A) as the center crossed the coast midway between Wrightsville and Topsail Beaches. The hurricane had been gradually weakening since its top speed of 115 mph on July 9 to 81 mph on July 11. Then, 12 hours before landfall, the estimated maximum 1 -minute winds increased to 104 mph. Bertha quickly dropped below hurricane strength when it moved inland over eastern North Carolina. Bertha then moved northeastward along the U.S. east coast, producing 45 to 60 mph sustained winds over land from northern North Carolina to New England. Over the nearby Atlantic, 70 mph winds were the rule. Bertha was declared extratropical on July 14 when the center moved from the Maine coast to New Brunswick, Canada. The extratropical storm brought 45 to 60 mph winds to the Canadian Maritime Provinces and was tracked to just south of Greenland on July 17. Meteorological Statistics Figures 2 and 3 show a plot, versus time, of the various data used to estimate the minimum central sea-level pressure and the maximum 1 -minute wind speed 33 feet above ground. Included are data from reconnaissance aircraft and satellite Dvorak-technique wind speed estimates. Appendix B, Table 2, lists selected surface observations of lowest pressure, peak wind, storm surge and rainfall values. Appendix B, Table 3, lists ship reports of 39 mph or greater that were associated with Bertha. The minimum pressure of 960 mb occurred at 2 a.m. EDT (0600 UTC) on July 9 and is based on a dropsonde measurement. The best track maximum sustained wind speed of 115 mph at the same time is based on a 700-mb flight-level wind speed of 140 mph, measured 22 miles east-northeast of the center. Observations are incomplete from the Leeward and Virgin Islands, but because the circular eyewall was 20 to 35 miles across, it is believed that hurricane conditions with sustained wind speeds to 86 mph could have occurred on Antigua, Barbuda, Nevis, St. Eustatius, St. Barthelemy, Anguilla, St. Martin, and from St. Thomas northward through the U.S. and British Virgin Islands. Experience with Hurricane Marilyn in 1995 suggests that even higher sustained winds can occur over mountainous terrain as is found on many of these islands. Winds 40 to 45 mph were experienced over portions of Puerto Rico as indicated by the San Juan observations in Appendix B, Table 2. A reconnaissance aircraft flight level wind speed of 127 mph in the northeast quadrant of the circulation several hours before landfall is the basis for estimating sustained surface winds of 104 mph on the coast at landfall. The lowest sea-level pressure observed at landfall was 977 mb at Surf City, North Carolina, and a value of 974 mb is assumed to be the minimum pressure at landfall. 7 1820 1819- 13M 13 loss - 9090 13 BEE] 67/83-W/L4.0% C32AT CTOW) 0 SAT C@@ #SAT CaPm) 03H ONJ T-NUM V Ac:src M 000130 X SURFAM as 956 - 0 L2 0 L2 0 12 6 12 0 12 0 12 0 12 8 L2 0 la 9 L2 8 12. ft L2 3 4 5 6 7 9 9 11 L2 13 14 DATE-TIME CUTC) Figure 2: Curve of minimum central sea-level pressure versus time. Uft 87/03-07/14/96 OZAT CTSM30 118 - 8 SAT CSARI@ *SAT CAFGWC) qo *3" 031 T-HUM Lee - V AC:SFC A AC: MT LUL .+bl 0 90 - OX ASCUIOZOMMPON lilt" 0 V A 13 76 - V 0 0 *0 60 - OD V z me 0 50 V 40 - 38 - 8 12 6 12 6 12 6 12 0 12 0 12 8 12 0 12 6 12 0 L2 0 12 8 L2 3 4 5 6 7 a 9 is 11 J.2 13 14 DATE-TIME (UTC). - Figure 3: Curve of maximum ]-minute speed versus time. 1. 8 Coastal storm surge flood heights, from Florida through New England, ranged from I to 4 feet, but values to 5 feet were estimated on the North Carolina coast from Cape Fear to Cape Lookout. A storm surge of 6 feet or a little higher was indicated near Swansboro, where 5 to 6 feet of water was "inside of businesses on the waterfront" (from Newport, North Carolina, National Weather Service Forecast Office Preliminary Storm Report). Values were forecast to be between 4-6 feet. Six tornadoes (four in Virginia, one in North Carolina and one in Maryland) have been confirmed during the passage of an outer rain band. Casualty and Damage Statistics There were 12 deaths related to Hurricane Bertha. Four in Florida included one from an evacuating military jet crashing into a house and three drownings from rip currents and high surf. Two deaths were reported in North Carolina (automobile accident and rip current drowning), while one surfer died in New Jersey. In Puerto Rico, one death occurred in an automobile accident and another died while surfing. On the French half of St. Martin, one person was electrocuted and another drowned after falling off a boat. The U.S. Virgin Islands, along with North Carolina, has been declared a Federal disaster area. Surveys indicate that Bertha damaged almost 2,500 homes on St. Thomas and St. John. For many, it was a second hit in the 10 months since Hurricane Marilyn devastated the same area. It is likely that there was beach erosion on the north coast of the Dominican Republic as Bertha passed to the north. The Bahamas were also affected by the weak side of the hurricane, but there are no damage figures available from either of these locations. The primary effects in North Carolina were to the coastal counties and mostly due to storm surge flooding and beach erosion, roof damage, piers washed away, fallen trees and damage to crops. A survey indicated more than 5,000 homes were damaged, mostly from storm surge. FEMA estimated 250,000 persons evacuated in South and North Carolina. Minor wind damage and flooding also spread along the path of the storm all the way to New England. The -Insurance Institute estimates $135 million in insured property damage, primarily along coastal North Carolina. A conservative ratio between total damage and insured property damage, compared to past land-falling hurricanes, is two to one. This would increase the total U.S. damage estimate to $270 million. No figures are available from the Caribbean. Warning and Forecast Critique Bertha moved on a fairly smooth track. The average official track forecast errors for Bertha ranged from 92 miles at 24 hours (32 cases) to 169 miles at 48 hours (29 cases) to 258 miles at 9 72 hours (27 cases). These errors are 15 percent or more lower than the previous 10-year averages of the official track errors and are from 15 to 40 percent lower than the CLIPER forecast errors for the same cases. Overall, the track model guidance also performed very well. However, the 8 p.m. EDT (0000 UTC) Aviation Model run on July 9, when Bertha was located just north of Puerto Rico, inexplicably showed the track recurving significantly further east than the previous run. All of the track guidance models that use the Aviation Model as a background environment also showed a similar track. This resulted in rather large official track forecast errors on July 9, with a 705-mile, 72-hour error on the 8 a.m. EDT (1200 UTC) forecast. The Aviation Model and some of the track guidance models recovered to an excellent forecast only 12 hours later. Fortunately, this guidance problem occurred three days prior to landfall in North Carolina and did not have a significant impact on U.S. warnings or on warnings for the Bahamas. Appendix B, Table 4, lists the various watches and warnings that were issued. Hurricane warnings were issued from Sebastian Inlet, Florida, to Chincoteague, Virginia, as well as for the Bahamas and for the islands of the northeastern Caribbean Sea from Antigua through Puerto Rico. Tropical storm warnings were issued from Sebastian Inlet to north of Deerfield Beach, Florida, and from north of Chincoteague to Watch Hill, Rhode Island. Almost all of the U.S. east coast was involved with some watch or warning which is the result of the storm track's expected close passage to the southeast U.S. coast. The hurricane watch for the North Carolina landfall area was issued 65 hours before landfall, and the hurricane warning was issued 47 hours before landfall. This is far more than the 36- and 24-hour lead times that the NHC strives for and is the result of the forward motion decreasing at a faster rate than expected. User Response For about 2 days, Bertha aimed at the north-central Florida east coast. During that time, all of the NWS track guidance models were predicting a turn to the north, near 76 degrees west. This was uncomfortably close to some of Florida's barrier islands and Cape Kennedy which have evacuation times exceeding 24 hours. At I I p.m. EDT on July 9 and again during the early morning hours of July 10, the State of Florida emergency management officials requested that NHC post a hurricane watch for the northern parts of their east coast, but NHC hurricane specialists felt that this was premature. On July 10, the governor issued mandatory evacuation notices at 10 a.m. EDT, and NHC put up hurricane warnings at 11 a.m. EDT for the state, north of Sebastian Inlet. Early that afternoon, Bertha began its recurvature toward the north in close agreement with the guidance models. Similarly, South Carolina state officials became concerned about possible nighttime evacuations on the evening of July 10 and were in great need of the latest information prior to the I I p.m. coordination call. The governor wanted to make a decision whether or not to order mandatory evacuations before the 11 p.m. news. Working and coordinating closely with NWSFO 10 Columbia, NHC was called at 7:30 p.m. EDT on July 10, requesting additional forecast track information. This request was received more than an hour before the hurricane specialist was to begin preparing the next forecast. The hurricane specialist and the HLT provided intermediate information on Bertha's location and strength, however, NHC could not release any track and intensity forecasts until the next NWS hurricane hotline call at 10 p.m. EDT, at which time the track and intensity forecasts would be coordinated and agreed upon. The HLT believed they kept state officials well informed regarding Bertha from an emergency management perspective. Hydrometeorological Prediction Center Meteorology of Heavy Rainfall Hurricane Bertha made landfall on the North Carolina coast on July 12 and was downgraded to a tropical storm shortly thereafter. The storm accelerated north-northeastward across coastal sections of the Mid-Atlantic and northeast United States prior to moving off the eastern New England coast late on July 13. Precipitable water values increased significantly to values in excess of 2 inches along the Eastern Seaboard as deep layered easterly flow developing in advance of the storm advected tropical moisture inland. Heavy rain occurred from the eastern Carolinas to Maine with many locations receiving 2- to 4-inch amounts (see Figure 4). Coastal North Carolina and southeast New York had isolated reports of over a half foot of rain. Amounts could have been much greater had Bertha's rapid motion not acted to limit rainfall duration. Guidance Products Primary guidance that aided HPC forecasters included varied output from numerical models (Geophysical Fluids Dynamical Lab [GFDL] model, Aviation Model [AVN], ETA Model [ETA], Nested Grid Model [NGM], Naval Operational Global Atmospheric Prediction System [NOGAPS], Medium-range Forecast Model [MRF], the MRF ensemble package, Medium-range Forecast Model Experimental [MRFX], European Center for Medium Range Weather Forecasting [ECMWF], United Kingdom Meteorological Office [UKMET], meso-ETA, NHC hurricane models, etc.), satellite imagery (Geostationary Operational Environmental Satellite [GOES]-8/9 Visible [VIS], Infrared [IR], Water vapor [WV], and Special Sensor Microwave Imagery [SSMI] microwave data), NEXRAD data, surface and upper-air observations and aircraft reconnaissance data. Model output was varied. While all models correctly indicated the potential for heavy rainfall over the eastern United States, the AVN forecast a Bertha track and associated quantitative precipitation forecast (QPF) that was shifted farther to the west than the ETA or NGM. The AVN track implied that a significant upslope rain event was imminent in the favored upslope region on the eastern side of Appalachian Mountains. The ETA and NGM supported a heavy rainfall event farther to the east. Weather pattern recognition and upper-air data analysis combined with trends from satellite imagery and radar to prompt HPC forecasters to prefer a solution close to but wetter than that depicted by the ETA. Subsequent observations deemed that our model assessments were on the right track. HPC staff produced an array of graphic products and accompanying narratives. These products provided an assessment of guidance and an excellent forecast of quantitative precipitation, flash flood potential and the sea level pressure pattern from current conditions up to five days (see Figures 5-7). Internal Coordination HPC forecasters participated in at least four regularly scheduled hurricane coordination calls each day that included verbal input from NHC, SPC, MPC, WFO's, WSH, SRH, ERH, Naval Atlantic Meteorology and Oceanography Center at Norfolk, Virginia, and the Naval Meteorology and Oceanography Facility at Jacksonville, Florida. Information discussed included Bertha's forecast strength and track from zero to five days and associated QPFs. Routine coordination calls were also performed with.FEMA. Marine Prediction Center Marine Meteorology As Bertha moved along the east coast recurving across the Maritimes and then well out to sea, it created significant marine meteorological conditions for an extensive segment of the MPC High Seas area of responsibility. The most notable conditions are chronicled below. 0 As Bertha passed near the Bahamas on July 10 at 8 a.m. AST (1200 UTC), well before the center entered MPC's area, 13-foot southerly swells had propagated as far north as 34N west of 69W. E On July I I at 8 a.m. AST (1200 UTC), while hurricane and tropical storm force winds remained south of 32N, the southern limit of the MPC forecast area, seas increased to 20 feet over the area south of 34N west of 70W. Near gale force winds extended out to 520 miles over the north semicircle. By July 12 at 8 a.m. EDT (1200 UTC), the center of Bertha had crossed 32N, approaching the coast of the Carolinas. Packing maximum sustained winds of 104 mph, the storm generated a maximum wave height estimated at 39 feet just south of Frying Pan Shoals. On July 13, Bertha moved inland across eastern North Carolina and southeast Virginia, weakening to a tropical storm with sustained winds of 52 mph and gusts to 63 mph at 8 a.m. EDT (1200 UTC). Winds near tropical storm force and 13-foot seas also occurred within 485 miles from the center. 12 TOTAL PRECIPITATION (inches) and PATH OF HURRICANE BERTHA 2 July 12 - 13, 1996 T. 4 4 2 4 41 CLIMATE PREDICTION CENTER 2 and NATIONAL HURRICANE CENTER Figure 4 13 25 CrQ 0 5 ct ct 50 2 2@ 50 ton 0129 25 00 OPRN CO z D129 2 4 H R t@ PEV I SE -24HR Y 2 END I GHR 0 FJ lz jrEND I NG \,Go SAT END ING \0 41. SUN 14 J 9 9 L- FCST SAT 13 96 @R FCSTR: FCSTR.- R, @2 j � By July 14 at 8 a.m. EDT (1200 UTC), Bertha had lost most of its tropical characteristics and was declared extratropical, becoming the sole responsibility of the MPC. Located along the New Brunswick coast, storm force winds to 63 mph and seas to 20 feet occurred within 495 miles over the southeast semicircle. Gale force winds and seas to 16 feet extended out to 620 miles over the southeast semicircle and 345 miles over the northwest semicircle. � On July 15 at 8 p.m. EDT (0000 UTC), former Bertha was downgraded from an extratropical storm to a gale, still packing winds to 52 mph with seas to 18 feet southeast of the center. � As the low mov *ed northeast across the open ocean, gale warnings were continued until July 17 at 2 a.m. EDT (0600 UTC). Through its final weakening stages as a gale from Newfoundland to just south of Cape Farewell, an extensive area of 13- to 21 -foot seas continued east and south of the center. Forecast and Guidance Products MPC forecast and analysis products are routinely broadcast to ships at sea. This is done through the U.S. Coast Guard high frequency communication facilities. The products consist of high seas text bulletins and graphic surface analysis and forecast products. In addition, voice broadcasts of warnings and forecasts which originate at the MPC are broadcast over station WWV, Boulder, Colorado. Although all of these products are produced according to a regular schedule, they all contain warning and forecast information as provided in the official TPC bulletin. The frequency of transmission for these products is every 6 hours and is based on the synoptic reporting times (0000 UTC, 0600 UTC, 1200 UTC, 1800 UTC). Coordination Coordination took place using the Hurricane Hotline or through telephone calls. TPC's conference calls were routinely monitored and telephone calls to TPC's Tropical Analysis and Forecast Branch were made for purposes of coordination. In anticipation of the system becoming extratropical and to cover expected conditions in offshore waters, MPC provided input for adjustments of TPC's tropical force wind/12-foot seas radius as the storm crossed north of 32N prior to making landfall along the southern North Carolina coast. The input was accepted and incorporated into the advisory. 15 Storm Prediction Center Forecast and Guidance Products Tornado watches associated with Bertha (see Appendix D). Tornado Watch issue (CDT) valid end 764 12/9:37 a.m. 10:00 a.m. 7:00 p.m. 765 12/1:36 p.m. 2:00 p.m. 7:00 p.m. 768 12/6:16 p.m. 7:00 p.m. 4:00 a.m. 770 12/11:41 p.m. midnight 7:00 a.m. 771 13/3:00 a.m. 3:15 a.m. 10:00 a.m. (replaced 768 and 770) 772 13/12:20 p.m. 12:45 p.m. 7:00 p.m. Mesoscale Meteorology and Coordination The SPC participated in all hurricane hotline discussions concerning Bertha during its lifetime, however, this was due to the fact that SPC forecasters were monitoring the hotline. NHC failed to roll call the SPC nor did it invite input from the SPC. NHC should include the SPC in its roll call list for coordinating advisories when a tropical cyclone threatens the U.S. mainland. During the 10 a.m. EDT coordination on July 12, the SPC forecaster commented that the tornado threat for the east coast was increasing and that he should probably issue a tornado watch (the first in a series). The hurricane specialist suggested that the SPC handle the details off-line with the individuals involved, and that he would mention -the possibility of tornadoes in the next bulletin. After the hotline call ended, the SPC forecaster called the offices in Raleigh, North Carolina, and Columbia, South Carolina, to coordinate the tornado watch (No. 764) which extended from northeastern South Carolina into southeastern and east-central North Carolina. During the course of the morning and early afternoon, several tornado warnings were issued in North Carolina with some occurring near the western and northern boundaries of the watch. Damage was reported with a thunderstorm near Raleigh. Since there was a strong surface high to the north (1026 mb center), the resulting pressure gradient was producing relatively strong low level flow (and shear) well to the north of the tropical storm center. Further, instability along and east of a surface trough extending from eastern North Carolina into southeastern Virginia had reached moderate values. These factors raised the concern that the tornado threat would be developing northward along and east of the trough line during the afternoon hours and that a watch would be needed northward from the original watch into southeastern Virginia. To make it simpler for the NWSFOs and NWSOs, the ending time 8 p.m. EDT (0000 UTC) was chosen to be the same as that with the original watch. The SPC forecaster called NWSFOs Raleigh, North Carolina, and Sterling, Virginia, to propose the new watch (No. 765). SPC mentioned that there was rotation with a cell on the Virginia/North Carolina border and that NWSO Wakefield, Virginia, had issued a tornado warning as a result. All agreed to the watch although Sterling was 16 a little hesitant. On several occasions, the SPC noted that it is apparent that during hazardous weather events when staffs at local offices are busiest, it can be difficult for them to stay abreast of evolving and immediate post-weather events in nearby NWSFOs/NWSOs areas of responsibility. Later in the evening, tornado watch Nos. 768 and 770 were issued as the tropical system tracked north. By early Saturday morning, July 13, the main east-west "spiral band" of isolated supercells which was over northern Virginia continued northward and had reached a northern Maryland/southern New Jersey line by around daybreak. Two rotating storms on the far western end of this activity were edging dangerously close to the western edge of the previous weather watch and nearing the eastern suburbs of Washington, D.C. Therefore, upon consultation with NWSFOs Sterling, Virginia; Philadelphia, Pennsylvania; and New York City, New York, the SPC forecaster decided to issue a new watch (No. 77 1) which covered a bit more of the "western shore" region of the Chesapeake Bay. Shortly after the watch was issued, a new thunderstorm began to rotate along the same spiral band which had produced the rotating storms east of Washington. This latter cell moved right across Baltimore and prompted a tornado warning from Sterling, but no tornadoes were reported. During the coordination call for Tornado Watch No. 77 1, NWSFO Philadelphia had no trouble accepting the watch as there had already been isolated rotating storms over Delaware Bay earlier. These storrns were well in advance of the Virginia/Maryland spiral band. On the other hand, NWSFO New York City really didn't want the watch (No. 77 1), but they were unaware of the fact that isolated supercells had been occurring as close to them as Delaware Bay earlier. There was good agreement with NWSFO Boston, Massachusetts, to the effect that any tornado threat in New England would be delayed for a while due to the presence of cooler/drier air near the surface (dew points were in the 60s vs. low to mid 70s in Maryland). As has been pointed out in several instances, the long standing procedures of watch coordination with field offices were, in general, effective. The coordination process provided the field with opportunities to contribute and, in most cases, coming to a consensus was not a problem. The SPC also noted the ability to dial into the WSR-88Ds that were being affected proved to be a valuable tool in assessing the tornado threat from Bertha. 17 Chapter 3 Regional Perspectives Southern Region 0 Puerto Rico and the U.S. Virgin Islands Data Collection WSR-88D Performance of the San Juan WSR-88D, an FAA radar, was acceptable as Bertha moved across the Virgin Islands and north of Puerto Rico. However, as of Saturday, July 6, the radar was inoperative with no apparent plans to return it to service. Representatives from SRH, WSH, Office of Systems Operation, and the Operational Support Facility spent several hours during the evening of July 6 coordinating and attempting to isolate the problem. On the morning of July 7, the FAA electronics technician (ET) had identified the problem as irregular fluctuations in the frequency of the generator used to power the RDA. Due to the remoteness of the site, the RDA does not have commercial power. The permanent fix to the generator's problem was to install an electronic governor on the generator, but this could not be done before Bertha's arrival. The FAA ET would attempt to temporarily stabilize the signal on a short-term basis but could not promise any definite success. The efforts were in fact successful as the WSR-88D remained operational for the duration of the event. Once the scope of the WSR-88D problem was made known to the appropriate FAA representatives, they were cooperative and restored the radar to service as quickly as possible. Automated Surface Observing System (ASOS) The ASOS units on St. Thomas and St. Croix functioned normally until Monday, July 8. On the morning of July 8, the control towers at both islands' airports were abandoned. FAA personnel have instructions to power down all equipment, including ASOS, when the towers are abandoned. Thus, no ASOS data were available as the storm impacted the islands, and no official wind or pressure data were available in real-time or after the event. The ASOS at San Juan remained operational, recording a maximum wind gust of 60 mph and a minimum pressure of 996.8 mb. 19 Warning and Forecast Services NWSFO San Juan began issuing HLSs at 11:20 p.m. AST, Saturday, July 6. The HLSs were issued at 6 hourly intervals through 5 p.m. on July 7, at which time they switched to 3 hourly statements. The statements contained a great deal of information presented in a segmented forrn as recommended by the Hurricane Marilyn Disaster Survey Report. Separate sections of the HLS were used to describe preparedness actions, the storm's position, storm surge, etc. This enabled customers to quickly find information regarding a particular aspect of the storm's threat. As Bertha approached and moved by Puerto Rico, HLSs were composed and disseminated every 2 hours. These HLSs were a marked improvement over those issued for Hurricane Marilyn, NOWs were prepared by NWSFO San Juan throughout the event. The NWSFO staff utilized information primarily from the WSR-88D when composing their NOWs. While the NOWs were informative, some of them were written in a style similar to the radar narrative summary (i.e., "At xxx pm ... radar indicated the center of Bertha..."). This had the effect of dating the products and focusing customers' attention to the storm's history rather than its forecast track and impacts. Dissemination and Coordination NWSFO San Juan communicated information regarding Bertha in a variety of ways. The primary d issernination means were NWR and NWWS. Local EMCs and media took an active role in relaying information to the citizens, using methods ranging from facsimile transmissions to live television and radio broadcasts. On St. Thomas, NWR broadcasts are provided through a service contract with radio station WAH. Normally, WAH provides NWR service between 6 a.m. AST and midnight. However, with the approach of Bertha, WAH conducted 24-hour operations from one day before the storm's arrival to two days after Bertha's impact. The St. Thomas office of the Virgin Islands Territory Emergency Management Agency (VITEMA) has the "state" NWWS drop provided by the NWS. The system was operational until mid-afternoon Monday, July 8, when power was lost to the VITEMA facility. The VITEMA-St. Thomas office was in frequent contact with NWSFO San Juan, as coordination calls were held at least five times daily starting on Friday, July 5. On St. Croix, VITEMA officials were still not receiving all NWWS products from the drop on St. Thomas. However, the St. Croix staff obtained the necessary information via coordination calls with the NWSFO San Juan, the Internet, The Weather Channel, and the VIFreenet. The VIFreenet is a free Internet-based service for Virgin Islands citizens. The service is maintained by private sources. VITEMA-St. Croix staff also gathered weather information from the FEMA Region I HF radio network. Although some problems were encountered with the NWWS, the VITEMA-St. Croix staff utilized multiple dissemination methods to receive information regarding Bertha. 20 Preparedness As in previous -years, the staff at NWSFO San Juan conducted an aggressive hurricane preparedness campaign. From October 1995 through June 1996, the office participated in 48 conferences, workshops, seminars, presentations and interviews related to hurricane preparedness. Among these activities was a 2-day workshop, conducted in cooperation with TPC/NHC, for VITEMA officials. The office also conducted an in-house hurricane seminar which was also attended by two members of the VITEMA-St. Croix staff. NWSFO continues their noteworthy preparedness efforts. Customer Response The local media was proactive, especially the print media outlets. The St. Croix Avis, the local newspaper on the island, covered Bertha as their headline story beginning on Saturday, July 6. NWS Public Information Statements documenting preparedness activities were printed in the Sunday and Monday (July 7 and 8) editions. VITEMA officials on both St. Thomas and St. Croix were satisfied with the infon-nation they received during the event, including the forecast track and intensity. Although Bertha was a fast-moving storm as it approached the islands, VITEMA staff and citizens felt they had ample time to prepare. Thus, the VITEMA staff believed that they received better services than during Hurricane Marilyn in 1995. Florida and Georgia Data Collection All of the data collection equipment in place along the Florida and Georgia coasts was operational during the event. The WSR-881), ASOS, upper air equipment, and data buoys performed well throughout Bertha's trip up the east coast. However, NWSFO Miami, Florida, noted a problem with a lack of data off the southeast Florida coast. Without the data, no wave information is available south of Melbourne, Florida. Wave heights can be estimated from local wind forecasts, but no observations of swells are routinely available and no "ground truth" wind/wave data can be accessed by forecasters. The NWSFO Miami WCM has talked to the Beach Patrol units in southeast Florida, asking them to call in rough surf reports to the office. This practice is not consistently done. Warning and Forecast Services All coastal offices issued frequent HLSs for their areas of responsibility. The HLSs roughly coincided with the release of advisory packages or intermediate advisories from TPC/NHC. 21 Because Bertha remained well away from the coastline, none of the Florida coastal offices needed to document the storm's progress in short term forecasts (NOW). During the evening of July 9, Bertha approached southeast Florida and TPC/NHC issued the Tropical Storm Warning from Deerfield Beach, Florida, northward. This was upgraded to a Hurricane Warning from Sebastian Inlet northward during the morning of July 10. NWSFO Miami, NWSO Melbourne, and NWSO Jacksonville, Florida, began issuing HLSs at around midnight on July 10. The HLSs were issued about every 3 hours until the warnings were discontinued (5 p.m. EDT on July 10 for Miami, 5 a.m. on July 11 for Melbourne, and 5 a.m. on July 12 for Jacksonville). Overall, the offices' HLSs were detailed and accurate. In cases where the HLSs were lengthy, a segmented approach was used. As at NWSFO San Juan, the Florida offices' segmented HLSs contained separate paragraphs discussing topics such as the storm itself, evacuation instructions, and tide information. It should be noted that all of Florida's coastal NWS offices issued timely and informative HLSs as Bertha passed offshore. In addition to the HLSs, NWSFO Miami issued heavy surf advisories and statements advising of rough surf conditions and potential strong rip currents through July 12. Unfortunately, one death occurred on July 10 due to a rip current off Jacksonville Beach. Later in the week, due to the lack of wave/swell data from off the southeast coast, the heavy surf advisory was prematurely canceled. Two more people were drowned and more than 100 were rescued along Dade County beaches on July 12 and 13 due to rip currents caused by swells from Bertha. The NWSFO Miami also issued a wind advisory on the morning of July 10 for inland portions of central and southern Florida with winds forecast to 35 mph. The wind advisory was extended to portions of north-central Florida during the late morning of July 10. Internal and External Coordination Although Bertha's meteorological impacts on the Florida and southeast Georgia coasts were minimal in the form of heavy surf and minor beach erosion, the storm was the subject of tense coordination between NHC and the State of Florida EMC. Because the NHC forecasters felt that the storm would turn to the north and eventually to the northeast, no watches or warnings were originally issued for the coast. By Tuesday, July 9, the storm still had not made an appreciable turn to the north. During a coordination with the state EMC on the evening of July 9, the state EMC virtually insisted that a watch or warning be issued so that their coastal county EMCs could initiate their hurricane plans. The storm still had not made an appreciable turn by the morning of July 10. During the 10 a.m. EDT conference call between NHC and the local offices, it was agreed that an upgrade to a hurricane warning was the best course of action. Bertha finally slowed and started to turn north by the afternoon of July 10, but the warnings were left up for precautionary reasons. In retrospect, the NHC official forecasts were accurate in depicting Bertha's decrease in forward speed and turn to the north. Although the turn did occur, the uncertainty involved with the forecast made coordination with the State of Florida EMC difficult. 22 The local offices along the coast were active in coordinating with their local emergency management officials. Staffs at the offices were involved with both the NHC coordination calls and the State of Florida conference calls as well. Offices utilized telephone and amateur radio communication when needed to elaborate on the information contained in HLSs. Customer Response Despite the coordination issues which occurred during Bertha's approach, local customers were satisfied in general with the services they received. The Palm Beach County EMC was supportive of the NWS warnings and forecasts and expressed displeasure only at the tenseness of the coordination between NHC and the state EMC. The Broward County EMC was concerned about the breakpoint chosen for the tropical storm warning (north of Deerfield Beach) since Deerfield Beach is in Broward County but the warning was for the area north of Broward County. Along the central Florida coast, there was some second-guessing of the warnings and evacuations. These concerns were especially evident in the local print media. However, the majority of the customers in the area felt that the actions taken were prudent considering the history and past track of the storm. The NWSO Melbourne WCM has engaged in ongoing meetings with local emergency managers to evaluate the communication and coordination issues. In northeast Florida and southeast Georgia, customers were satisfied as well with the services from NWSO Jacksonville, Florida. The regional media were centrally located in the Duval County EMC. On the afternoon of July 10, the NWSO Jacksonville MIC relocated to the Duval County, Florida, EMC to field media inquiries and to enhance coordination with Duval County and other local emergency management officials. Community response in the coastal areas was immediate upon issuance of the hurricane warning. NWSO Jacksonville received a letter of appreciation from the Camden County, Georgia, EMC, and several other local agencies gave verbal compliments to the office. As in east-central Florida, the NWSO Jacksonville WCM and SOO have engaged in frequent evaluation meetings with local customers. Eastern Region Impacts Hurricane Bertha was an unusual early season Category 2 hurricane which made landfall near Wrightsville Beach, North Carolina. Bertha was the first significant hurricane to directly affect the state between Wilmington and Morehead City since Hurricane Donna in 1960. The American Insurance Association estimates that $135 million of insured property losses occurred during Bertha with $ 100 million having occurred in North Carolina, $25 million in 23 Virginia, and $10 million in South Carolina. Claims represent insured losses resulting from factors other than tidal inundation (i.e., mostly wind). The hardest hit area was just to the east of the eye at landfall ... Topsail Beach and Swansboro, North Carolina. Water was 5 to 6 feet deep in waterfront businesses in Swansboro. Highest wind gusts reported on land were 108 mph at Camp Lejeune and Jacksonville and 95 mph at Wrightsville Beach (all in North Carolina). Wind gusts of 80 mph were observed at NWSO Newport, North Carolina. In Wilmington, North Carolina, just to the west of the eye, winds gusted to 70 mph. Over water, the Frying Pan Shoals C-Man tower, 35 miles off the North Carolina coast, registered a gust to 115 mph. NWSO Wilmington, North Carolina, recorded 5.66 inches of rain during Bertha's passage. In North Carolina, moderate to severe amounts of beach erosion occurred in areas of Bertha's greatest impact. Mason Inlet lost 50 feet of land, leaving the inlet that much closer to Shell Island Resort. At Emerald Isle, the dunes dropped vertically 8 to 10 feet at the erosion line. At Kure Beach, there was 4 feet of vertical erosion against the boardwalk, leaving the strand narrower by 40 feet. On the Outer Banks, more than 150,000 people (mostly tourists) evacuated, back to the mainland. Further south, well over 100,000 people evacuated to the west of 1-95 and out of harms way. Outside the areas of greatest impact, Bertha's effects were much less dramatic. In South Carolina, the highest measured wind was at the downtown Charleston Handar with a peak gust of 58 mph. Higher peak gusts likely occurred along the barrier island beachfronts. No significant beach erosion occurred as the maximum tide departure occurred near the time of low tide. Rainfall was generally between I and 2 inches. Widespread power outages occurred over North Carolina with as many as 250,000 customers without power in the eastern portions of the state. About 80,000 people were evacuated in South Carolina and 20,000 in Georgia. Revenue losses due to evacuations of tourists and residents were estimated to be about $40 million. One person drowned off Myrtle Beach in advance of the storm while attempting to, surf the large waves and swells. In Virginia, four tornadic storms associated with Bertha resulted in much of the damage across eastern portions of the state. The strongest tornado struck Northumberland County, Virginia, and resulted in nine injuries. Rainfall totals ranged from 3 to 6 inches on the Virginia Capes and the lower Maryland shore to 2 to 4 inches over inland sections of Virginia east of Interstate 95. Mostly minor to locally moderate tidal flooding and beach erosion occurred. Extensive power outages occurred over eastern Virginia, with as many as 115,000 customers without power in the eastern sections of the state. In the NWSO Wakefield, Virginia, county warning area (CWA), nearly 1,800 people were evacuated. Inland portions of Delaware and Maryland experienced heavy rain and 'gusty winds from what was now Tropical Storm Bertha although the most significant impact was from several weak tornadoes. There was some attendant wind damage associated with the stronger thunderstorms in 24 Bertha's rain bands. The Patuxent River, Maryland, Naval Air Station recorded a gust of 48 mph and 4.03 inches of rain. Rainfall totals ranged ftom 4 to 4.5 inches along the western shore of the Chesapeake Bay and I to 2.5 inches across metropolitan Washington, D.C. In Bertha's aftermath, about 2,000 homes were without power in northern Virginia, 5,000 around the District of Columbia and surrounding Maryland suburbs, and 10,000 in and around Baltimore. In New Jersey, the most significant impact from Bertha was the heavy rain with 2- to 4-inch storm totals common in coastal sections, with a few isolated reports greater than 6 inches. One such report was 6.69 inches in Estelle Manor, New Jersey. Generally, coastal and flash flooding were not a problem although moderate wave action was responsible for one surfing fatality. Harvey Cedars, New Jersey, reported a 63-mph gust with some downed trees and resultant power outages which were a consequence of the area's experiencing strong gusty winds and saturated soils. On Long Island, winds gusted to 64 mph on Babylon Village, New York. Rainfall totals were as high as 4.65 inches at Pomona in Rockland County, New York. In New York City and Long Island, rainfall amounts ranged from 0.76 inches in Quogue to 2.49 inches at Mt. Sinai (both on Long Island). Mostly minor to locally moderate beach erosion occurred. In southern New England, most of Bertha's adverse effects were minimal and resulted in only minor coastal flooding. One strong band of convection, however, resulted in strong winds and numerous power outages across Rhode Island and southeast Massachusetts. The highest reported gust was 87 mph in New Bedford, Massachusetts. The maximum reported rainfall was 7.20 inches at Billerica, Massachusetts. Rainfall totals exceeded 5 inches at several locations in northern Rhode Island, northeast Connecticut and interior parts of eastern Massachusetts. The remainder of New England essentially had to deal with the locally heavy rains associated with Bertha's remnants. Total rainfall across the area was generally in the 2- to 4-inch range with lower amounts in northern Maine and northern New Hampshire. There were several reports of water over the roads in the areas covered by flood warnings, but no significant problems or damage was reported. (See Appendix E for the Aerial Survey.) Warning and Forecast Services Timely hurricane watches and warnings were posted commensurate with the threat posed by Bertha. However, intensity forecasts prior to landfall in North Carolina were not on target. Information from NHC indicated that Hurricane Bertha had weakened as she approached the North Carolina coast. The hurricane had been gradually weakening since its top speed of 115 mph on July 9 to 81 mph on July 11. Then 12 hours before landfall, the estimated maximum 1 -minute winds increased to 104 mph. Information relayed to eastern North Carolina emergency managers mirrored information received from the NHC. Forecast wind speeds, especially inland, were too low. 25 Forecasts of sound-side flooding of the Outer Banks were based on a numerical model developed at North Carolina State University. The forecast based on the late afternoon advisory on July I I called for a 6- to 8-foot surge. This forecast did not take into account decreasing wind speeds as Bertha moved north through eastern North Carolina. As a result, the surge forecast was too high. Actual surges were on the order of I to 2 feet. South of this area from Cape Lookout to Cape Fear, the storm surge was between 5 and 6 feet and was forecast to be between 4 and 6 feet. Some lower Maryland Eastern Shore emergency managers expressed concern with the lack of a Hurricane or Tropical Storm Watch north of the Hurricane Warning area. It was their feeling that a Watch would have afforded them the freedom to take action commensurate with the level of threat they perceived existed. Further north, local NWSFO and NWSO products were excellent and accurately depicted the hazards posed by the now weakening tropical storm. Early on in the event, there was some initial forecast disagreement across southern New Hampshire. This was subsequently resolved by a joint telephone call with the New Hampshire Office of Emergency Management by staffs at NWSFOs at Taunton, Massachusetts, and Gray, Maine. Generally, the accurate and timely NWS warnings and forecasts allowed customers both on land and water to adequately prepare for Bertha. Internal and External Coordination Internal NWS and external coordination during Bertha's passage through Eastern Region went well in most instances. Timely notification to Federal, state, county and local agencies were initiated by local NWS offices well in advance of the official watches and warnings. This ensured that users were well-prepared days in advance of the hurricane's impact. External coordination was accomplished primarily via telephone. Some users in the heart of Bertha's path obtained information through a bulletin board system at NWSO Newport, North Carolina. The primary backup system at the NWSO, amateur radio operators, was used to gather information but was not needed for coordination purposes since the telephone system did not fail. Internal NW`9 coordination was accomplished primarily via telephone, the hurricane hotline, and AFOS. Some user complaints were also received regarding the utility of the inland winds model. As noted earlier, this largely resulted from the forecast of strong inland winds in the NHC marine advisory not reflecting observed conditions. In northeast North Carolina, the hurricane revealed some coordination problems with regard to the timely dissemination of important weather information between NWSO Wakefield, Virginia, and the emergency managers of the coastal counties within Wakefield's CWA. The MIC and 26 SOO met with the emergency managers and resolved these issues. Additionally, emergency management officials stated they continue to have tight budgets and cannot afford NWWS. EMWIN was emphasized while the use of Internet and the bulletin board service was mentioned. In North Carolina, some problems were noted regarding NHC initiated calls to the EMC. When a hurricane warning is posted for the North Carolina coast, the NWSFO Raleigh MIC and WCM brief the state EMC in person when each new advisory is issued. State emergency management officials also call the NWSFO as appropriate, between visits by the MIC or WCM, to get intermediate updates on the hurricane's progress. NHC briefing calls frequently came at times when state officials were already in the midst of a briefing by the NWSFO or with local emergency management officials along the coast. At a number of NWS offices, HAM radio SKYWARN volunteers received and passed along information. In at least one instance (NWSO Wilmington, North Carolina), Packet Radio was their primary communication device. Local emergency management officials with Packet Radio often received the hurricane advisories and local office statements more than a half hour before they were received by other means. Additionally, the Packet Radio also provided them with Doppler (WSR-88D) radar images from the radar coded message product. Data Collection and Communication The coastal portion of the NWSO Morehead City, North Carolina, CWA is covered with data collection devices ranging from ASOS sites, Datalogger, and DARDC wind equipment. Most units failed once power was lost. Amateur radio operators were used to gather information throughout the storm and proved to be NWSO Morehead City's most reliable data-gathering source. Generally, though, the sparsity of land and over water.real-time observations proved once again to be an obstacle as Bertha slammed onshore and traversed through eastern North Carolina. Local NWS office work on fine-tuning the WSR-89D rainfall estimates continued during Bertha. With the passage of each tropical system through the region, more radar data are helping local offices to fine tune the most accurate reflectivity (Z-R) relationship to utilize. This will aid NWSFO/NWSO assessments of the (rainfall) flooding potential during future tropical system passages. WSR-88D level H, Ell, and IV data were archived by NWS offices in Bertha's path. Unfortunately, the archive level II data at NWSO Morehead City was inoperative throughout the event. In addition, many NWSFOs and NWSOs archived upper level plots, surface observations and model gridded data during the hurricane. In North Carolina, the New Bern and Cape Hatteras NWR failed for a time as the hurricane was reaching peak strength. This is a problem which the Region is attempting to minimize for future 27 hurricanes and other severe weather episodes through the use of NWS-owned and operated uninterruptible power supply and back-up generator equipment at NWR transmitter locations. A number of offices noted a truncation of some of the more lengthy HLS on TWC. Moreover, it was also noted that some of the NOWs were cut off at six, as opposed to eight lines. Obviously, this premature truncation of the HLS and NOW will result in need-to-know information not being passed on to the viewer and as such needs to be resolved with TWC. A telephone company problem led to the NWR's Blue Hill transmitter, which serves the Greater Boston area, being off the air for several hours during the height of the storm. Additionally, the Providence ASOS temperature and dew point sensor went out of service during Bertha's passage through the area. User Response User response was quite good. The successful evacuations and low casualty rates reflect the growing sophistication and team efforts of the NWS and the emergency management community although a number of false perceptions as pertains to (he hurricane's perceived threat linger. Emergency management personnel in North Carolina are well trained and made the appropriate evacuation decisions. All used either the "HurrTrak" or "Hurrevac" computer software to aid in their decision-making process. Conference calls that included all emergency managers as well as NWSO Morehead City and NWSO Wakefield personnel were initiated by the North Carolina Area A Coordinator (includes northeastern North Carolina counties). Occasional conference calls were initiated by the Area C Coordinator (central and southeastern North Carolina counties). The media and public responded quite well. Most tourists evacuated when ordered, however, the majority of the year-round residents remained on the North Carolina barrier islands. Further north, the states of Maryland and Virginia were contacted after the event to assess their reaction. Feedback from the state level was strongly positive. Some concern was expressed by one emergency manager on the Maryland Eastern Shore, regarding the NWS use of one of the hurricane break points during the event. Just to the south of Bertha's greatest impact, individuals who were not allowed to return to evacuated areas immediately after the hurricane and subsequently saw only limited damage noted that they would not evacuate when the next hurricane approached. This attitude is a continuing problem that the NWS and FEMA will have to resolve in present and future educational outreach efforts. 28 Preparedness The true measure of a warning program's effectiveness is the degree of response that the warning elicits from the public and local officials. The NWS must work closely with emergency managers, officials and the media to gain their trust and to ensure that their requirements for technical information are met. Similarly, active public awareness campaigns conducted with local officials and the media foster heightened awareness of hazards presented by hurricanes. Numerous prefearly season hurricane preparedness activities were conducted throughout the Region prior to Bertha making landfall in North Carolina and then moving up the east coast, a short distance inland. Most individuals knew how to respond to Bertha's threat, and thereby demonstrated that NWS community preparedness/outreach efforts indeed paid off. It was interesting to note, however, that North Carolina Preparedness Week was scheduled in North Carolina for the week following Bertha's passage. Needless to say, the week's activities were canceled in lieu of the "real-life" experiences associated with this early season hurricane. In South Carolina, working relations with local and state emergency managers are in excellent condition thanks to regular South Carolina Hurricane Task Force meetings and work accomplished through the local NWSO Charleston and NWSFO Columbia preparedness programs. Similar actions at other Eastern Region coastal NWSFOs/NWSOs translated to successful local preparedness efforts during the days leading up to Bertha's passage through the Region. During state conference calls, the state and local officials displayed reasonable expectations and asked pertinent questions of the local office NWS participants. These emergency managers seemed to understand well the role local NWSFO and NWSO offices play and used NWS- provided information to their full advantage. Finally, Eastern Region NWSFOs and NWSOs were well prepared internally for Bertha with appropriate staffing changes initiated well in advance of the hurricane/tropical storm. This ensured that local NWS offices were properly staffed to meet the challenges posed by this hurricane. Externally, shelters were opened where needed, and again, no preparedness problems were noted. 29 Appendix A Saffir-Simpson Hurricane Scale* Category Definition-Likely Effects ONE Winds 74-95 mph: No real damage to building structures. Damage primarily to unanchored mobile homes, shrubbery, and trees. Also, some coastal road flooding and minor pier damage. TWO Winds 96-110 mph: Some roofing material, door, and window damage to buildings. Considerable damage to vegetation, mobile homes, and piers. Small craft in unprotected anchorages break moorings. THREE Winds 111-130 mph: Some structural damage to small residences and utility buildings with a minor amount of curtainwall failures. Mobile homes are destroyed. Flooding near the coast destroys smaller structures with larger structures damaged by floating debris. Terrain may be flooded well inland. FOUR Winds 131-155 mph: More extensive curtainwall failures with some complete roof structure failure on small residences. Major erosion of beach areas. Major damage to lower floors of structures near the shore. Terrain may be flooded well inland. FIVE Winds greater than 155 mph: Complete roof failure on many residences and industrial buildings. Some complete building failures with small utility buildings blown over or away. Major damage to lower floors of all structures located near the shoreline. Massive evacuation of residential areas may be required. In operational use, the scale corresponds to the I -minute average sustained wind speed as opposed to gusts which could be 20 percent higher or more. A-1 Appendix B t Meteorological Tables B-I Table 1. Best track, Huff icane Bertha, 5 - 14 July, 1996 (updated 4 August 1996), Date/Time Position Pressure Wind Speed Stage (UTC) (mb) (kt) Lat. ('N) Lon. (*W) 0510000 9.8 34.0 1009 30 tropical depression 0600 10.2 36.3 1008 30 11 1200 11.0 39.0 1007 35 tropical storm 1800 12.0 41.2 1006 35 06/0000 12.7 43.9 1005 35 0600 13.1 46.6 1004 35 1200 13.7 48.7 1002 40 1800 14.2 51.0 1000 45 07/0000 14.9 52.9 999 50 0600 15.6 54.8 997 55 1200 16.4 56.9 995 60 1800 16.5 58.4 992 70 hurricane 08/0000 17.0 60.1 988 75 0600 17.5 61.8 985 75 1200 18.0 63.5 983 70 1800 18.6 64.9 978 75 09/0000 19.4 66.1 970 80 0600 20.3 67.7 960 100 1200 21.4 69.4 965 100 1800 22.5 71.1 967 90 10/0000 23.6 72.6 969 85 0600 24.5 74.0 971 80 1200 25.4 75.3 968 80 1800 26.4 75.8 966 80 11/0000 27.5 76.4 968 75 0600 28.3 76.8 972 75 1200 29.2 77.5 977 75 1800 30.0 78.0 980 70 12/0000 30.7 78.3 982 70 0600 31.2 78.6 984 70 1200 32.2 78.4 975 85 1800 33.6 78.1 974 90 13/0000 35.0 77.6 993 65 0600 36.7 77.0 993 60 tropical storm 1200 38.3 76.1 994 60 1800 40.2 74.5 994 60 14/0000 42.1 71.9 994 60 0600 44.1 69.0 995 55 1200 46.0 66.0 995 50 extratropical 1800 47.0 62.0 995 50 15/0000 48.0 57.0 995 50 0600 49.0 52.0 996 45 1200 51.0 47.0 996 40 1800 54.0 44.0 996 40 16/0000 57.5 42.5 991 40 0600 58.5 42.5 988 40 1200 59.5 .42.0 988 45 1800 59.8 41.0 985 45 17/0000 60.0 40.0 993 40 0600 60.5 39.0 1001 35 09/0600 20.3 67.7 960 100 minimum pressure 12/2000 34.3 77.8 974 90 landfall* midway between Wrightsville and Topsail Beaches *The eyewall also passed over Antigua, Barbuda, St. Barthelemy, Anguilla, St Martin, St Thomas, and the British Virgin Islands and passed close to a number of other islands in the northeastern Caribbean Sea. Table2. Hurricane Bertha selected surface observations. Julv 1996. Press. Date/time Sustained Feak Date/time I Storm Storm total Location (mb) (UTQ wind(kt)a gust (UTO' ge(ft)c tide(ft)" rain(in.) _Lsu-rge(ft)c Florida Jacksonville 1011.9 j 11/2256 22(2 min) 27 11/1721 trace Jacksonville Beach 2 Georgia Sea Island 3 St. Simon 3 South Carolina Charleston international airport 1008.5 1 12/1155 30 (2-min.) 39 12/0130 1.48 Charleston city office 36 50 12/1130 0.94 Charleston harbor 2.7 Cheraw 0.38 Cherry Grove pier 68 Garden City pier 54 Loris 2.77 Myrtle Beach 2.40 Myrtle Beach, Sands Resort 4.73 Myrtle Beach Springmaid pier 3.3 Myrtle Beach pavilion 52 Summerville 1.40 North Carolina Beaufort Duke marine lab 2.8 56 12/2125 Beaufort Bath 6 Bellhaven 7 Bnmswick T -5 6 Carolina Beach 64 12/2242 1 Cherry Point (NKT) East Wilmington 70 Greenville 76 4.11 63 13/0100 Ratteras ferry ofice Kure Beach 1 49 180 12/1835 -T- Lake Waccamaw 2.39 Newport 994.3 178 2.95 12/2208 4.56c New Bem 162 New River (NCA) 94 12/2021 Pongo River 4.5 Seymour Johnson AFB 986.3 12/2355 1 33 52 12/2155 4.03 Snow Hill 5.44 S. Pamlico River 3 Southport (Nixon) 978.3 2/1835 55 74 12/1703 Surf City (Horodner) 977 1 12/2005 Swansboro Williamston 4.10 Wilmington (ILM) 978.7 12/2028 46(2-min) 61 12/1902 i 1.3 5.66 Wilmington NC state port 77 i Wilmington port terminal 980.1 12/1850 NOAA ship Whiting, 3.4.2N 77-96W Cape Lookout Buoy 62 112/2300 Diamond Shoals Buoy 171 113/0200 977.5 12/1800 7700 min) 101 12/1610 FPSN7 (CMAN. 80 ft asi) 80 Wrightsville Bch. Banks ch. Virginia 1311155 Andrews AFB (ADW) 1000.6 35 13/0750 0.94 Baltimore Washington airport 11000.3 1 13/1349 141 13/0704 2.28 Cape Charles 61 13/0330 Eastvillc, Northhampton Co. 7.00 Table2(continued). Hurricane Bertha selected surface observations. JuIv 1996. Press. Dateitime Sustained Peak Dateitime Storm Storm total Location (Mb) (UTC) wind(kt)c gust (UTO, surge(ft)g tide (ft)' rain(in.) Virginia Elizabeth City (EC6) 998.9 13/1313 39 i48 13/0055 Newport News (PHF) 1 993.1 1310750 Norfolk International airport. 995.1 13/0756 31 42 13/0740 2.44 Norfolk NAS (NGU) 994.4 13/0755 30 42 13/0155 Norfolk, Sewells Point 2.5 Oceana NAS (NTU) 1 997.2 13/0755 Pasquotank River -7 5 Patuxent Fiver NAS (NHK) 1 995.0 13/1100 42 13/0603 1 4.03 Plantation Creek 7.20 Potomac river at Wisc. Ave. 5 Wallops Island ---T-42 47 13/0605 Washington Dulles airport 29 13/1218 1.12 Washington National airport 1 1001.7 1 13/1221 1.24 29 13/1346 'i 1 5.80 Willis Wharf, Accornac Co. Chesapeake Light Stn. Buoy 49 56 13/0700 Fenwick Is. Buoy (44009) 41 48 13/1300 Maryland Nanticoke, Wicontico Co. 5.03 Salisbury 5.62 Delaware Dover 995.6 1 1311155 30 50 Lewes 1.58 New Jersey Atlantic City 2.27 2.41 Estelle Manor 6.59 Harvey Cedars 55 Newark airport 25 (2-min.) 35 13/1521 -Teterboro airport 26 (2-min.) 38 13/1547 River Vale 2.17 West Milford 4.85 Pennsylvania Philadelphia 29 36 12.43 New York Ambrose Light Tower buoy 48 Babylon ViUage, L.I. 64 Buoy southwest of Fire Is. 37 Brookhaven airport 26 36 13/1947 East Quogue, Suffolk co. 1 0.76 -JFK airport 31(2-min.) 37 113/1458 LaGuardia airport 29 (2-min.) 38 113/1518 2.49 Mt. Sinai(New York City) Ossining, Westchester Co. 3.09 Pomona, Rockland Co. 4.65 Connecticut 0.91 13/1547 Bridgeport airport 27 34 Orange Preston 148 Shelton 14.06 Vernon 5.50 Table2(continued). Hurricane Bertha selected surface observations. Julv 1996. Date/time Sustained e/time Storm Storm total Location Press. (UTC) windi gust (UM Tsurgetffll tide(ft)' rain(in.) _I Peak Dat Rhode Island Fox Point hurricatie barrier 2.1 Sachuest Point (Middletown) 54 i64 14/00109 hand-held anemometer Providence 995.9 13/2336 Maine Goshen 1 5.70 Billerica 7.20 Boston 996.3 14/0056 1 New Bedford 45 76 13/2030 1'.7 Taunton ASOS 995.6 14/0041 Puerto Rico San Juan 996.8 08/2056 39 1 49 08/2056 4 averaging period is I min unless otherwise indicated. 'Date/time is for sustained wind when both sustained and gust are listed. ' Storm surge is water height above normal astronomical tide level. dStorm tide is water height above NGVD. 'Top of rain gage blew off and "a lot of rain was sucked out". Table 3. Ship reports of 34 knots or higher wind speed, associated with Hurricane Bertha, July 1996. date/time ship name latitude longitude wind dir/ pressure (UTC) (ON) (1,W) speed(knots) (mb) 07/1200 FNOR 17.1 53.8 110/39 1014.2 08/1800 WZJE 21.6 64.3 130/35 1013.0 09/0000 WCHF 21.6 64.7 080/036 1009.1 09/0300 WZJE 20.0 64.6 100/38 1007.0 09/0300 WCHF 21.0 65.0 100/48 1008.9 09/0600 WZJE 16.9 65.1 130/40 1007.5 09/0600 WCHF 20.6 65.1 130/48 1006.9 09/0900 WCHF 20.4 65.3 130/42 1009.0 09/1800 WZJF 23.0 67.9 100/43 1012.0 09/1800 P3TA4 24.3 67.0 110/34 1013.2 11/1800 KIRF 28.5 78.8 270/35 1006.0 11/1800 C6YC 29.6 73.5 150/040 1020.5 11/1800 KRPP 30.8 79.9 050140 1010.0 11/1800 PCAM 31.6 80.4 040/43 1010.0 11/1800 DQFT 32.4 72.6 140/35 1022.5 11/2100 C6YC 28.9 74.3 140/42 1018.8 12/0000 C6YC 28.3 74.8 140/38 1017.5 12/0000 KRPP 30.8 80.1 030/38 1004.7 12/0000 PCAM 31.0 80.9 020/47 1010.0 12/0600 DQFT 30.7 74.7 140/58 1019.0 12/0600 PCAM 31.1 80.6 360/66 1005.0 12/0600 KRLZ 31.3 76.5 130/45 1008.0 12/0900 KRLZ 30.8 76.4 150155 1007.3 12/1200 P3HR2 28.6 79.7 270/34 1015.5 12/1200 KRLZ 30.4 76.4 150/43 1008.8 12/1200 ELSD9 30.9 75.4 150/37 1016.0 12/1200 SHIP 34.9 71.1 180/52 12/1500 DQFT 30.0 77.3 180/39 1015.0 Table 3(continued). Ship reports of 34 knots or higher wind speed, associated with Hurricane Bertha, July 1996. date/time ship name latitude longitude wind dir/ pressure (UTC) (ON) (0 W) speed(knots) (mb) 12/1800 DQFT 30.1 78.1 210/47 1015.5 12/1800 9VKR 30.5 74.8 170/35 1013.0 12/1800 ELSD9 30.7 76.3 150/37 1017.5 12/1800 SHIP 32.7 73.6 160/40 1016.8 12/1800 9VBK 33.8 73.9 140/34 1017.8 '13/0000 WCjY 32.7 75.2 180/35 1022.2 13/0000 9VBK 33.6 75.1 170/38 1012.2 13/0600 3FWD 32.2 73.7 180/37 1017.0 13/0600 9VBK 33.3 76.0 200/38 1013.3 13/1200 4XGS 39.3 73.9 140/60 1006.0 13/1200 9VUM 37.9 73.2 150/40 1008.1 13/1500 3FEB5 36.5 69.9 180/40 1019.6 9VUM 37.6 73.9 210/44 1006.5 13/1500 4XGS 38.7 74.1 160/60 1000.0 13/1800 3FEB5 36.0 70.0 170/37 1 1017.9 13/1800 9VUM 37.3 74.8 240/62 1009.6 13/1800 4XGS 38.1 74.3 210/40 1005.0 13/1800 LENA3 40.6 72.4 170/40 1004.0 13/2100 DDQS 39.5 69.0 170/37 1012.6 14/0000 DDQS 39.6 69.5 210/47 1009.8 14/0000 NJPJ 41.9 70.3 180/35 1 1006.1 14/0600 DDQS 39.7 70.4 240/35 1013.3 14/0900 VOLG 45.0 60.4 170/35 611.4 14/1200 JKPS 37.4 66. 230/39 1022.0 14/1200 CG2683 43.3 62.6 230/50 1007.0 14/1200 3EXP6 43.9 61.3 180/48 1007.8 14/1200 VOLG 45.2 59.4 170/35 Table 4. Watch and warning summary, Hurricane Bertha. July 1996. Date/time Action Location (UTC) 06/1500 hurricane watch Antigua. Barbuda, Nevis. Montserrat. St Kitts, Anguilla. Saba, St Eustatius, Dominica, and Dutch St Maarten 06/2100 hurricane watch Guadeloupe, St Barthelemy, French St Martin, U.S. and British Virgin Islands 07/0300 hurricane warning Dominica northward to Anguilla and St Maarten 07/1200 hurricane watch Puerto Rico 07/1500 hurricane warning U.S. and British Virgin Islands and Puerto Rico 08/0000 tropical storm watch Dominican Republic from Isla Saona to Cabo Frances Viejo 08/0300 hurricane watch Dominican Republic from Isla Saona to Cabo Frances Viejo 08/0900 hurricane watch Turks and Caicos Islands 08/1200 hurricane warning Dominican Republic from Cabo Caucedo, to Monte Cristo, southeastern Bahamas 08/1500 hurricane warning discontinued Dominica 08/1800 hurricane watch Haiti from St Nicolas to border of Dominican Republic 08/1800 huricane warning discontinued Leeward Islands south and east of St Eustatius 08/2100 hurricane warning Turks and Caicos Islands and southeastern Bahamas 08/2100 hurricane warning discontinued Leeward Islands 08/2100 hurricane watch central Bahamas 09/0100 hurricane warning discontinued U.S. and British Virgin Islands 09/0300 hurricane warning discontinued Puerto Rico 09/0300 tropical storm warning Haiti from St Nicolas to border of Dominican Republic 09/0900 hurricane warning central Bahamas 09/0900 hurricane watch northwestern Bahamas 09/1500 watches and warnings discontinued Dominican Republic and Haiti 09/2100 hurricane warning northwestern Bahamas 10/0300 tropical storm warning north of Deerfield Beach. FL to Brunswick, GA 10/0300 hurricane watch north of Brunswick to NCNA border including Pamlico and Albemarle Sounds Table 4(cont.). Watch and warning summary, Hurricane Bertha. Julv 1996. Date/time Action Location (UTQ 10/0900 hurricane warning discontinued Turks and Caicos Islands and southeastern Bahamas 1011500 hurricane warning Sebastian Inlet, FL to Cape Romain. SC 10/1800 hurricane warning discontinued central Bahamas 10/2100 hurricane warning Cape Romain to NCNA border including Pamlico and Albemarle Sounds 10/2100 tropical storm warning discontinued south of Sebastian Inlet. FL 11/0300 hurricane watch NCNA border to Chincoteague VA including southern Chesapeake Bay 11/0600 hurricane warning discontinued northwestern Bahamas 11/0900 hurricane warning discontinued south of Brunswick, GA 12/0900 hurricane warning discontinued Savannah, Ga southward 12/1500 tropical storm warning NCNA border to Chincoteague, VA including southern Chesapeake Bay 12/1900 hurricane warning discontinued Cape Romain, SC southward 12/2100 hurricane warning NCNA border to Chincoteague, VA including the Hampton Roads area 12/2100 tropical storm warning north of Chincoteague, VA to Watch Hill, RI including the lower Delaware Bay 12/2100 tropical storm watch east of Watch Hill to the Merrimack River. MA 13/0300 hurricane warning discontinued south of Topsail Beach, NC 13/0300 hurricane watch discontinued lower Chesapeake Bay 0/0300 tropical storm warning all of Chesapeake Bay and the lower tidal Potomac River and all of Delaware Bay 13/0700 hurricane warning to tropical storm Topsail Beach, NC to Chincoteage, VA including warning Albemarle and Pamlico Sounds 13/0900 hurricane warning discontined NCNA border southward 13/1200 tropical storm warning discontinued south of Fenwick Island, DE 13/1500 tropical storm warning discontinued south of Brigatine, NJ and Delaware Bay 13/2100 tropical storm warning discontinued south of Fire Island, NY 14/0000 tropical storm warning discontinued south of Watch Hill, RI 14/0300 tropical storm warning discontinued remainder of US east coast Appendix C Fujita Tornado Intensity Scale Cate2ory Definition-Effective (FO) Gale tornado (40-72 mph): Light damage. Some damage to chimneys; break branches off trees; push over shallow-rooted trees; damage sign boards. (Fl) Moderate tornado (73-112 mph): Moderate damage. The lower limit is the beginning of hurricane wind speed; peel surface off roofs; mobile homes pushed off foundations or overturned; moving autos pushed off the roads. (F2) Significant tornado (113-157 mph): Considerable damage. Roofs torn off frame houses; mobile homes demolished; boxcars pushed over; large trees snapped or uprooted; light-object. missiles generated. (F3) Severe tornado (158-206 mph): Severe damage. Roofs and some walls torn off well-constructed houses; trains overturned; most trees in forest uprooted; heavy cars lifted off ground and thrown. (F4) Devastating tornado (207-260 mph): Devastating damage. Well-constructed houses leveled; structure with weak foundation blown off some distance; cars thrown and large missiles generated. (F5) Incredible tornado (261-318 mph): Incredible damage. Strong frame houses lifted off foundations and carried considerable distance to disintegrate; automobile sized missiles fly through the air in excess of 100 yards; trees debarked; incredible phenomena will occur. C-1 Appendix D Storm Prediction Center Tornado Watch Areas D-1 TORNADO #0764 Issue Time: 1437 UTC 07/12/1996 Valid Time: 1500 UTC 07/12/1996 End Time: 0000 UTC 07/13/1996 ... . . ....... Area: 19244 sq. mi. ...... ... Issuing Forecaster: JOHNS ZCZC MKCSAW4 ALL 130000;332,0800 360,0774 2160,0751 332,0773; WWUS40 KMKC 121437 CDR MKC AWW 121437 CDR WW 764 TORNADO SC NC AND ADJ CSTL WTRS 12150OZ 13000OZ AXIS.30 STATUTE MILES EAST AND WEST OF LINE.. 20SSE MYR/MYRTLE BEACH SC/ - 70NNE EWN/NEW BERN NCI -AVIATION COORDS.. 60NM E/W /65SE FLO - 24SW ECG/ WIND GUSTS.30 KNOTS. MAX TOPS TO 550. MEAN WIND VECTOR 12035. NNNN ZCZC MKCSEL4 ALL 130000;332,0800 360,0774 360,0751 332,0773; WWUS9 KMKC 121437 COR MKC WW 121437 CDR SCZ000-NCZ000-1 30000- BULLETIN - IMMEDIATE BROADCAST REQUESTED TORNADO WATCH NUMBER 764 NATIONAL WEATHER SERVICE KANSAS CITY MO 1037 AM EDT FRI JUL 12 1996 A ... THE STORM PREDICTION CENTER HAS ISSUED A TORNADO WATCH FOR PARTS OF SOUTH CAROLINA PARTS OF NORTH CAROLINA AND ADJACENT COASTAL WATERS EFFECTIVE THIS FRIDAY MORNING AND EVENING UNTIL 800 PM EDT. TORNADOES ... LARGE HAIL ... DANGEROUS LIGHTNING AND DAMAGING THUNDERSTORM WINDS ARE POSSIBLE IN THESE AREAS. <NO CANCELLATION> THE TORNADO WATCH AREA IS ALONG AND 70 STATUTE MILES EAST AND WEST OF A LINE FROM 20 MILES SOUTH SOUTHEAST OF MYRTLE BEACH SOUTH CAROLINA TO 70 MILES NORTH NORTHEAST OF NEW BERN NORTH CAROLINA. REMEMBER ... A TORNADO WATCH MEANS CONDITIONS ARE FAVORABLE FOR TORNADOES AND SEVERE THUNDERSTORMS IN AND CLOSE TO THE WATCH AREA. PERSONS IN THESE AREAS SHOULD BE ON THE LOOKOUT FOR THREATENING WEATHER CONDITIONS AND LISTEN FOR LATER STATEMENTS AND POSSIBLE WARNINGS. C ... TORNADOES AND A FEW SVR TSTNIS WITH HAIL SFC AND ALF TO EXTRM TURBC AND SFC WIND GUSTS TO 70 KNOTS. A FEW CBS WITH MAX TOPS TO 550. MEAN WIND VECTOR 12035. D... E... OTIR TSTMS... ...JOHNS NNNN TORNADO #0765 Issue Time: 1836 UTC 07/12/1996 Valid Time: 1900 UTC 07/12/1996 End Time: 0000 LITC 07/13/1996 Area: 13139 sq. mi. -Ij Issuing Forecaster: JOHNS ZCZC MKCSAW5 ALL 130000;375,0742 360,0751 360,0774 375,0765: WWUS40 KMKC 121836 MKC AWW 121836 WW 765 TORNADO VA NC AND ADJ CSTL WTRS 1219DOZ - 13000OZ AXIS..70 STATUTE MILES EAST AND -WEST OF LINE.. 70NNE ORF/NORFOLK VA/ - 25SW ECG/ELIZABETH CITY NC/ -AVIATION COORDS.. 60NM E/W /32SSW SBY - 21SW ECG/ WIND GUSTS..65 KNOTS. MAX TOPS TO 530. MEAN WIND VECTOR 12030. NNNN ZCZC MKCSEL5 ALL 130000;375,0742 360,0751 360,0774 375,0765; WWUS9 KMKC 121836 MKC WW 121836 VAZ00O-NCZ0OO-1 30000- BULLETIN - IMMEDIATE BROADCAST REQUESTED TORNADO WATCH NUMBER 765 NATIONAL WEATHER SERVICE KANSAS CITY MO 236 PM EDT FRI JUL 12 1996 A ... THE STORM PREDICTION cr=NtER HAS ISSUED A TORNADO WATCH FOR PARTS OF SOUTHEASTERN VIRGINIA PARTS OF NORTHEASTERN NORTH CAROLINA AND ADJACENT COASTAL WATERS EFFECTIVE THIS FRIDAY AFTERNOON AND EVENING UNTIL 800 PM EDT. TORNADOES ... LARGE HAIL ... DANGEROUS LIGHTNING AND DAMAGING THUNDERSTORM WINDS ARE POSSIBLE IN THESE AREAS. <NO CANCELLATION> THE TORNADO WATCH AREA IS ALONG AND 70 STATUTE MILES EAST AND WEST OF A LINE FROM 70 MILES NORTH NORTHEAST OF NORFOLK VIRGINIA TO 25 MILES SOUTHWEST OF ELIZABETH CITY NORTH CAROLINA. REMEMBER ... A TORNADO WATCH MEANS CONDITIONS ARE FAVORABLE FOR TORNADOES AND SEVERE THUNDERSTORMS IN AND CLOSE TO THE WATCH AREA. PERSONS IN THESE AREAS SHOULD BE ON THE LOOKOUT FOR THREATENING WEATHER CONDITIONS AND LISTEN FOR LATER STATEMENTS AND POSSIBLE WARNINGS. C ... TORNADOES AND A FEW SVR TSTMS WITH EXTRM TURSC AND SFC WIND GUSTS TO 65 KNOTS. A FEW CBS WITH MAX TOPS TO 530. MEAN WIND VECTOR 12030. D ... WIND FIELDS ARE INCRG NWD ALG SFC TROF THRU NERN NC AND SERN VA WHERE AMS IS MODERATE. TORNADO PTNL WL BE INCRG NWD WITH TIME INTO SERN VA. E...OTR TSTMS... CONT..WW 764.. ... JOHNS NNNN n TORNADO #0768 Issue Time: 2316 UTC 07/12/1996 Valid Time: 0000 UTC 07/13/1996 End Time: 0900 UTC 07/13/1996 Replace Time: 0815 UTC Q7/13/1996 Area: 27993 sq. mi. Issuing Forecaster: WIESS ZCZC MKCSAWB ALL 130900;3",0780 382,0771 382,0743 3",0753; WWUS40 KMKC 122316 MKC AWW 122316 WW 768 TORNADO NC VA MD AND ADJ CSTL WTRS 13000OZ - 13090OZ AXIS..70 STATUTE MILES EAST AND WEST OF LINE.. 25SSE EWN/NEW BERN NCI - 25WNW SBY/SALISBURY MD/ _AVIATION COORDS.. 60NM E/W /57NE ILM - 20WNW SBY/ HAIL SURFACE AND ALOFT.. 1/2 INCHES. WIND GUSTS..70 KNOTS. MAX TOPS TO 500. MEAN WIND VECTOR 17050, REPLACES WW 764 .. SC NC AND ADJ CSTL WTRS REPLACES WW 765..VA NC AND ADJ CSTL WTRS NNNN ZCZC MKCSEL8 ALL 130900;344,0780 382,0771 382,0743 344,0753; WWUS9 KMKC 122316 MKC WW 122316 NCZOOO-VAZOOO-MDZOOO- 130900- BULLETIN - IMMEDIATE BROADCAST REQUESTED TORNADO WATCH NUMBER 768 NATIONAL WEATHER SERVICE KANSAS CITY MO 716 PM EDT FRI JUL 12 1996 A ... THE STORM PREDICTION CENTER HAS ISSUED A TORNADO WATCH FOR PARTS OF EASTERN NORTH CAROLINA PARTS OF SOUTHEAST VIRGINIA PARTS OF SOUTHEAST MARYLAND AND ADJACENT COASTAL WATERS EFFECTIVE THIS FRIDAY NIGHT AND SATURDAY MORNING UNTIL 500 AM EDT. TORNADOES ... LARGE HAIL ... DANGEROUS LIGHTNING AND DAMAGING THUNDERSTORM WINDS ARE POSSIBLE IN THESE AREAS. <NO CANCELLATION> THE TORNADO WATCH AREA IS ALONG AND 70 STATUTE MILES EAST AND WEST OF A LINE FROM 25 MILES SOUTH SOUTHEAST OF NEW BERN NORTH CAROLINA TO 25 MILES WEST NORTHWEST OF SALISBURY MARYLAND. REMEMBER ... A TORNADO WATCH MEANS CONDITIONS ARE FAVORABLE FOR TORNADOES AND SEVERE THUNDERSTORMS IN AND CLOSE TO THE WATCH AREA. PERSONS IN THESE AREAS SHOULD BE ON THE LOOKOUT FOR THREATENING WEATHER CONDITIONS AND LISTEN FOR LATER STATEMENTS AND POSSIBLE WARNINGS. B. OTHER WATCH INFORMATION.. THIS TORNADO WATCH REPLACES TORNADO WXTCH NUMBER 764...TORNADO WATCH NUMBER 765. WATCH NUMBERS 764 ... 765 WILL NOT BE IN EFFECT AFTER 800 PM EDT. C ... TORNADOES AND A FEW SVR TSTMS WITH HAIL SFC AND ALF TO 1/2 INCHES. EXTRM TURBC AND SFC WIND GUSTS TO 70 KNOTS. A FEW CBS WITH MAX TOPS TO 500. MEAN WIND VECTOR 17050. 0 ... BANDS OF CNVTN ASSOCD WITH HURCN BERTHA EXPCD TO CONT ROTG NWO AND NWWD THRU THE NERN QUAD OF THE CRCLN CENTER. IR STLT IMAGERY INDCS COOLING CLD TOPS NR THE EYEWALL RGN AS WELL AS IN THE OFFSHORE FEEDER BANDS. 50-70 KT LOW LVL WINDS ABV THE SFC EAST OF THE CENTER WILL MAINTAIN VERY STG LOW LVL SHEAR/HELICITY FVRBL FOR THREAT OF ISOLD TORNADOES AS BERTHA MOVES NWD TNGT. E ... OTR TSTMS... THIS WATCH REPLACES..WW 764..WW 765.. CONT..WW 766..WW 767., ... WEISS NNNN TORNADO #0770 Issue Time: 0441 UTC 07/13/1996 Valid Time: 0500 UTC 07/13/1996 End Time: 1200 UTC 07/13/1996 Replace Time: 0815 UTC Q7/13/1996 Area: 10567 sq. mi. Issuing Forecaster: WEISS ZCZC MKCSAWO, ALL 131200;382,0765 401,0755 401,0734 382,0744; WWUS40 KMKC 130441 CDR MKC AWW 130441 COR WW 770 TORNADO MD DE NJ PA AND ADJ CSTL WTRS 13050OZ - 13120OZ AXIS..55 STATUTE MILES EAST AND WEST OF LINE.. 50SSW DOV/DOVER DE/ - TTNfTRENTON NJ/ -AVIATION COORDS.. 45NM E/W /16WNW SBY - 39SW EWR/ HAIL SURFACE AND ALOFT.. 1/2 INCHES. WIND GUSTS..70 KNOTS. MAX TOPS TO 450. MEAN WIND VECTOR 17040. NNNN ZCZC MKCSELO ALL 131200;382,0765 401,0755 401,0734 382,0744: WWUS9 KMKC 130441 CDR MKC WW 130441 CDR MDZOOD-DEZOOO-NJZOOO-PAZOOO-131200- BULLETIN - IMMEDIATE BROADCAST REQUESTED TORNADO WATCH NUMBER 770 NATIONAL WEATHER SERVICE KANSAS CITY MO 1241 AM EDT SAT JUL 13 1996 A ... THE STORM PREDICTION CENTER HAS ISSUED A TORNADO WATCH FOR PARTS OF MARYLAND PARTS OF DELAWARE PARTS OF NEW JERSEY PARTS OF PENNSYLVANIA AND ADJACENT COASTAL WATERS EFFECTIVE THIS SATURDAY MORNING UNTIL 800 AM EDT. TORNADOES LARGE HAIL ... DANGEROUS LIGHTNING AND DAMAGING THUNDERST E POSSIBLE N'THESE AREAS. <NO CANCELLATION> THE TORNADO WATCH AREA IS ALONG AND 55 STATUTE MILES EAST AND WEST OM 50 MILES SOUTH SOUTHWEST OF DOVER DELAWARE TO TRENTON NEW JERSE REMEMBER ... A TORNADO WATCH MEANS CONDITIONS ARE FAVORABLE FOR TORN ERE THUNDERSTORMS IN AND CLOSE TO THE WATCH AREA. PERSONS IN THESE A D BE ON THE LOOKOUT FOR THREATENING WEATHER CONDITIONS AND LISTEN FOF TEMENTS AND POSSIBLE WARNINGS. C ... TORNADOES AND A FEW SVR TSTMS WITH HAIL SFC AND ALF TO 112 INCHES. TURBC AND SFC WIND GUSTS TO 70 KNOTS. A FEW CBS WITH MAX TOPS TO 450. ND VECTOR 17040. D... E...OTR TSTMS... CONT..WW 768..WW 769..WW 770.. ... WEISS NNNN TORNADO #0771 Issue Time: 0800 LITC 07/13/1996 Valid Time: 0815 UTC 07/13/1996 End Time: 1500 LITC 07/13/1996 X.N, Area: 41048 sq. mi. Issuing Forecaster: CORFIDI ZCZC MKCSAW1 ALL 131500;372,0781 411,0751 411,0712 373,0743; WWUS40 KMKC 130800 MKC AWW 130800 WW 771 TORNADO VA DC MD PA NJ NY CT DE AND ADJ CSTL WTRS 130815Z 13150OZ AXIS.. 100 STATUTE MILES EAST AND WEST OF LINE.. 50E RIC/RICHMOND VA/ - 15NW BDR/BRIDGEPORT CT/ -AVIATION COORDS.. 85NM E/W /37NNW ORF - 13NW BDR/ WIND GUSTS-70 KNOTS. MAX TOPS TO 450. MEAN WIND VECTOR 17040. REPLACES WW 768 .. NC VA MO AND ADJ CSTL WTRS REPLACES WW 770..MD DE NJ PA AND ADJ CSTL WTRS NNNN ZCZC MKCSEL1 ALL 131500:372,0781 411,0761 411,0712 373,0743; WWUS9 KMKC 130800 MKC WW 130800 VAZOOO-DCZOOO-MDZOOO-PAZOOO-NJZOOO-NYZOOO-CTZOOO-DEZOOO-1 31500- BULLETIN - IMMEDIATE BROADCAST REQUESTED TORNADO WATCH NUMBER 771 NATIONAL WEATHER SERVICE KANSAS CITY MO 400 AM EDT SAT JUL 13 1996 A.-THE STORM PREDICTION CENTER HAS ISSUED A TORNADO WATCH FOR PARTS OF EASTERN VIRGINIA DISTRICT OF COLUMBIA PARTS OF CENTRAL AND EASTERN MARYLAND DELAWARE PARTS OF EASTERN PENNSYLVANIA NEW JERSEY PARTS OF SOUTHEASTERN NEW YORK PARTS OF SOUTHWESTERN CONNECTICUT AND ADJACENT COASTAL WATERS <NO CANCELLATION> EFFECTIVE THIS SATURDAY MORNING UNTIL 1100 AM EDT. TORNADOES ... LARGE HAIL ... DANGEROUS LIGHTNING AND DAMAGING THUNDERSTORM WINDS ARE POSSIBLE IN THESE AREAS. THE TORNADO WATCH AREA IS ALONG AND 100 STATUTE MILES EAST AND WEST OF A LINE FROM 50 MILES EAST OF RICHMOND VIRGINIA TO 15 MILES NORTHWEST OF BRIDGEPORT CONNECTICUT. REMEMBER-A TORNADO WATCH MEANS CONDITIONS ARE FAVORABLE FOR TORNADOES AND SEVERE THUNDERSTORMS IN AND CLOSE TO THE WATCH AREA. PERSONS IN THESE AREAS SHOULD BE ON THE LOOKOUT FOR THREATENING WEATHER CONDITIONS AND LISTEN FOR LATER STATEMENTS AND POSSIBLE WARNINGS. B..OTHER WATCH INFORMATION.. THIS TORNADO WATCH REPLACES TORNADO WATCH NUMBER 768 ... TORNADO WATCH NUMBER 770. WATCH NUMBER 768 ... 770 WILL NOT BE IN EFFECT AFTER 415 AM EDT. C ... TORNADOES AND A FEW SVR TSTMS WITH HAIL SFC AND ALF TO EXTRM TUROC AND SFC WIND GUSTS TO 70 KNOTS. A FEW CBS WITH MAX TOPS TO 450. MEAN WIND VECTOR 17040. D ... THREAT FOR ISOLD TORNADOES LKLY TO CONT INVOF FEEDER BANDS E/NE OF BERTHA ... WHERE SFC DWPTS HAVE RECOVERED TO ADA 70F AND LOLVL SPD/DIREC SHEAR RMNS VRY STG. E... OTR TSTMS... THIS WATCH REPLACES..WW 768..WW 770.. ... CORFIDI TORNADO #0772 Issue Time: 1720 UTC 07/13/1996 Valid Time: 1745 UTC 07/13/1996 End Time: 0000 UTC 07/14/1996 Area: 15600 sq. mi. Issuing Forecaster: VESCIO ..... . .. ... L ZCZC MKCSAW2 ALL 140000;404,0731 424,0722 424,0693 404,0702; WWUS40 KMKC 131720 MKC AWW 131720 WW 772 TORNADO MA CT RI NY AND ADJ CSTL WTRS 131745Z - 14000OZ AXIS '* 75 STATUTE MILES EAST AND WEST OF LINE.. 3OSSW BID/BLOCK ISLAND RI/ - 25N BOS/BOSTON MA/ -AVIATION COORDS.. 65NM E/W /59SSW PVD - 38SE CON/ HAIL SURFACE AND ALOFT.. 1/2 INCHES. WIND GUSTS..70 KNOTS. MAX TOPS TO 450. MEAN WIND VECTOR 20030. NNNN ZCZC MKCSEL2 ALL 140000;464,0731 424,0722 424,0693 404,0702; WWUS9 KMKC 131720 MKC WW 131720 MAZOOO-CTZOOO-RIZOOO-NYZOOO-140000- BULLETIN - IMMEDIATE BROADCAST REQUESTED TORNADO WATCH NUMBER 772 NATIONAL WEATHER SERVICE KANSAS CITY MO 120 PM EDT SAT JUL 13 1996 A ... THE STORM PREDICTION CENTER HAS ISSUED A TORNADO WATCH FOR PARTS OF CENTRAL AND EASTERN MASSACHUSETTS PARTS OF CENTRAL AND EASTERN CONNECTICUT RHODE ISLAND PARTS OF EASTERN LONG ISLAND NEW YORK AND ADJACENT COASTAL WATERS EFFECTIVE THIS SATURDAY AFTERNOON AND EVENING UNTIL 800 PM EDT. TORNADOES ... LARGE HAIL ... DANGEROUS LIGHTNING AND DAMAGING THUNDERSTORM WINDS ARE POSSIBLE IN THESE AREAS. <NO CANCELLATION> THE TORNADO WATCH AREA IS ALONG AND 75 STATUTE MILES EAST AND WEST OF A LINE FROM 30 MILES SOUTH SOUTHWEST OF BLOCK ISLAND RHODE ISLAND TO 25 MILES NORTH OF BOSTON MASSACHUSETTS. REMEMBER ... A TORNADO WATCH MEANS CONDITIONS ARE FAVORABLE FOR TORNADOES AND SEVERE THUNDERSTORMS IN AND CLOSE TO THE WATCH AREA. PERSONS IN THESE AREAS SHOULD BE ON THE LOOKOUT FOR THREATENING WEATHER CONDITIONS AND LISTEN FOR LATER STATEMENTS AND POSSIBLE WARNINGS. C ... TORNADOES AND A FEW SVR,TSTMS WITH HAIL SFC AND ALF TO 1/2 INCHES. EXTRM TURBC AND SFC WIND GUSTS TO 70 KNOTS. A FEW CBS WITH MAX TOPS TO 450. MEAN WIND VECTOR 20030. D ... CNVTN IS INCRG JUST S OF LONG ISLAND ON E SIDE OF T.S. BERTHA. THIS ACTVTY IS XPCD TO LIFT NWD DURG THE AFTN AND EARLY EVE. DRY INTRUSION NOTED ON LATEST WATER VAPOR IMAGERY MAY ALLOW FOR SUFF AMS DESTABILIZATION FOR A FEW SVR TSTMS PRODUCING DAMAGING WNDS AN[ ISOLD TORNADOES. ... VESCIO NNNN Appendix E Aerial Survey Introduction Hurricane Bertha struck the North Carolina coast on the afternoon and evening of July 12, 1996. 1 arrived in Wilmington, North Carolina, on Monday, July 15. During the afternoon of July 15, a flight was made on the NOAA Aircraft Operations Center (AOC) DeHavilland Twin Otter. An aerial flight was made along the coastline from Cape Fear to Cape Lookout, North Carolina. This 3 1/2 hour flight was spent examining wind damage along the coast and intercostal waterway area. On Tuesday, July 16, no flight was scheduled as U.S. Geological Survey was flying on a storm surge mission. A ground survey of agricultural damage was conducted in the afternoon from Wilmington northward to just south of Warsaw, North Carolina. Wednesday, July 17, a trip was made with the MIC and WCM of NWSO Wilmington to observe the damage at Carolina Beach and Kure Beach, North Carolina. Both locations lost piers. There were some structures with portions of roofs torn off, but most structures survived well. In the afternoon, another flight was made with the Twin Otter. This flight concentrated on inland damage from the hurricane. Damage was mainly confined to crops and trees and extended as far north as Goldsboro, North Carolina. Coastal Damage Wind damage was visible from the air from just north of Bald Head Island near Cape Fear to Atlantic Beach, south of Morehead City, North Carolina. The most intense damage with peak wind gusts estimated at 110 mph was located between North Topsail Beach and Emerald Isle, North Carolina. In this area, many beach homes suffered roof damage. The roof damage varied from torn off shingles to whole roofs lost. There were very few total failures to structures due to wind. Of those that did fail, it appears they were modular home or mobile home structures. Visible coastal erosion was found, especially to beach sand dunes and to a coastal road at North Topsail Beach. However, the damage was not as severe as in other recent landfalling hurricanes, such as Opal in 1995. Further inland along the edge of the intercoastal waterway, the primary damage was to trees. Numerous trees fell with most falling in the area northeast of the eye from west of Surf City northeastward to just east of Swansboro, North Carolina. This area extended northward to the E-1 southern portions of Jacksonville. Wind gusts in this area were estimated to be between 80 and 100 mph. Inland Agricultural Areas Agricultural areas were particularly exposed to high winds from Hurricane Bertha. Much of the damage was to corn and tobacco crops. Both of these crops are vulnerable to high wind gusts. Much of the agricultural damage was west of the eyewall rather than east. This is due in part to much of the corn and tobacco crops being grown further west rather than along the eastern coastal areas. The storm tracked slowly northward along the North Carolina and South Carolina coastline before making landfall. Thus, crops were exposed to strong northeast winds for an extended period of time, and this caused crops to weaken significantly. The agricultural damage extended as far as 40 miles west of the eye and as far north as Goldsboro, North Carolina, which is approximately 75 to 80 miles inland. Most of the damage occurred in unprotected (no blockage by trees or buildings) areas along the edge of fields. Some fields had more widespread damage where a streak of particularly high wind toppled several rows of corn or tobacco. Tornadoes The initial SPC log of tornadoes associated with Bertha indicated at least six touched down. Three of these reported tornadoes were investigated by air. There was no visible evidence of tornado damage in the three locations of New Bern, Newport, and near Winnabow, North Carolina. It is possible that some of these tornadoes either blended in with the hurricane damage or were not visible from the air. East of Jacksonville, North Carolina, a tornado track was identified from the air and was rated an Fl on the Fujita Tornado Intensity Scale (Appendix Q with a path length of approximately 1 1/2miles. The one area that a tornado most likely did not occur was near Winnabow. Conclusions and Acknowledgments The most severe wind damage associated with Hurricane Bertha was along the coast and east of the eye center, typical of most hurricanes. The peak wind speeds estimated from the aerial survey was 110 mph. Although damage to structures was minimal in this hurricane, agricultural damage was extensive. The author would like to thank the AOC and their fine pilots, Lieutenants Pickett and Longenecker. The author would also like to extend thanks to the Office of the Federal Coordinator for their help in making this survey possible and to Richard Anthony and Tom Matheson from the NWSO Wilmington, North Carolina, for their assistance in conducting the survey. E-2 *U.S. GOVERNMENT PRINTING OFFICE: 1997 418-592/40031 I I I I -1 1 3 6668 14100 0689 f, DATE DUE GAYLORDINO. 2333 1 PRINTED IN USA -