EXECUTIVE SUMMARY

Tropical Storm Alberto originated in Senegal on June 18, 1994, as a tropical wave. The system became Tropical Depression One on June 30 at approximately 0600 Universal Coordinated Time (UTC). On July 2 at about 0000 UTC, the depression strengthened in the Gulf of Mexico near the Yucatan Peninsula to become Tropical Storm Alberto. When the center made landfall near Destin, Florida, at 1500 UTC on July 3, Alberto was at its peak intensity, 993 millibars and 55-knot winds. Winds then quickly subsided, and Alberto's central pressure rose rapidly.

After landfall, the motion of the storm slowed and precipitation increased. The storm moved slowly through Alabama into Georgia, stalling just south of Atlanta. Over the next few days it reversed its course and then looped back on its previous course before ultimately dissipating. During that period it dumped copious amounts of rain across the area. Amounts as high as 21.1 inches in 24 hours were observed at Americus, Georgia. The Weather Surveillance Radar-1988 Doppler (WSR-88D) provided the forecasters with a very good representation of the areal extent of the precipitation although it underestimated precipitation amounts somewhat. This rainfall produced record and near-record flooding along the Flint, Ocmulgee, Chattahoochee, Choctawhatchee, and Apalachicola Rivers. Overall, flash flooding and flooding caused by the rainfall from Alberto took 33 lives, destroyed thousands of homes (including some entire communities), forced approximately 50,000 people to be evacuated, and caused property damage (including lost crops) estimated as high as $750 million.

Based on the current technologies available to the National Weather Service (NWS) offices in the area affected by Alberto, the offices in general performed their forecast and warning functions in an exemplary manner. The NWS received high praise for its products and services from all affected parties (emergency managers, the media, and the general public). Throughout Georgia, Alabama, and Florida, the disaster survey team found a remarkably universal degree of high regard.

The lack of negative comments may be attributed in part to the fact that external perceptions and expectations of NWS's present and future capabilities were quite limited relative to what the NWS believes it can and should be able to do as a result of the modernization now underway. Initial uncertainties regarding Alberto's landfall, the failure to predict that the storm would stall over Georgia instead of moving to the northeast, the Bainbridge forecast discrepancy, and the relatively short lead-time of some flash flood warnings are all examples of where users of NWS forecasts and products should, in the future, be able to expect a more accurate and timely service.

Many of the deaths in this event can be attributed to individual lack of judgment: refusal to evacuate despite the request of emergency managers and other authorities, attempts to either drive around barricades or on inundated roads, and other actions obviously inappropriate in the face of the hazard. Approximately two-thirds of the deaths were related to vehicular incidents. They also represent a small fraction (less than 0.1 percent) of the total number of people evacuated. Nevertheless, the high loss of life is troubling and clearly leaves room for improvement. A few people were seemingly unaware of the impacts of their weather-related decisions. Some complacency was evident early in the event. A factor which may have contributed to this complacency is the fact that recent floods in the Southeast have not been nearly as severe as this one. Therefore, most residents did not have previous experience in dealing with such a dangerous flood event.

Although general external impressions of NWS performance were favorable, the disaster survey team's closer examination revealed a number of causes for concern and opportunities for improvement (a summary of all the findings and recommendations of the disaster survey team is located in Appendix A):

- Perhaps as much as any single factor, the inability of NWS centralized model guidance to predict the reversal of the storm motion and precipitation amounts significantly limits the utility of NWS products and services. The NWS should work with the Office of Atmospheric Research and other Federal agencies, as well as the academic research community, to improve such predictions.

- The forecast for the Flint River at Bainbridge received considerable media and public attention when the river crested well below the forecasted level. The forecast at Bainbridge needs to be investigated and appropriate changes made to the forecast scheme.

- There is some indication that a number of problems occurred during the event that were related to the public's perception of the interfaces between responsibilities. Examples of these interfaces include the following: between the National Hurricane Center and National Meteorological Center's Meteorological Operations Division as the storm made landfall; between Weather Service Forecast Offices; between Weather Service Forecast Offices and Weather Service Offices; between Weather Service Forecast Offices and the River Forecast Center; and between the NWS and the media, emergency managers, and the general public. Interfaces are inevitable. Modernization in the NWS will cause a shift in a number of them. The NWS should develop an inventory of particularly important interfaces and ensure that the treatment of these receive special attention and priority.

- Transition to the modernized NWS and its associated staffing configurations pose special challenges. Throughout the affected region, offices had to deal with added stress in their handling of the event by conflicts between scheduled training and urgent operations, by recent introduction of the new technologies (particularly the WSR-88D), and by vacancies. The NWS should reexamine its approach to staffing and training during the transition with an eye to the special vulnerabilities represented by extreme events and make necessary adjustments. In addition, the NWS should continue working with management at the Department of Commerce, the National Oceanic and Atmospheric Administration (NOAA) , and the Office of Management and Budget , and with the Congress, to ensure that this is the last such wrenching modernization the NWS undertakes. In the future, modernization must be a continuing process, not a disruptive event.

- Preparedness is a special issue. As warnings and forecasts of particular events improve, opportunities for saving lives and property will depend increasingly on preparedness. In the modernized NWS, it will be challenging for a smaller number of offices to work with communities and other affected parties spread over large geographical areas to build the needed relationships and coordination on an ongoing basis. The NWS should identify resources for improving the capabilities of Weather Service Forecast Offices and future Weather Forecast Offices to build community preparedness with special focus on taking advantage of the "information highway."

- New demands on NOAA for information are created by increasing and changing societal vulnerability to weather, growing awareness of this vulnerability, and technological advances, especially in computing and communications. NOAA should continue to shift emphasis from particular events to ongoing processes of preparedness. It should create national capabilities that parallel the Federal Emergency Management Agency's capabilities for special emergency response and disaster relief operations. NOAA should also give more emphasis to the development of all-hazard telecommunications capability for NOAA Weather Radio.

---