[From the U.S. Government Printing Office, www.gpo.gov]
NOAA NDSR 73-1 A UNITED STATES- DEPARTMENT OFDiatrS ve PUBLICATION Natural DiatrSr~yReport 73-1 CoastalZoneU.S. DEPARTMENT OF COMMERCE Information 'National Oceanic and Atmospheric Administration, COASTAL ZONE ~~~~~~~~~~~~~~~~~~~~~~ANES 4 ~~~~~~ ~~~~~~~~~~~~~~~~AGNES JUNE20 -'~~~~~~~~~~~~~~~~~~~~~~GE \J) -~ 44 4 A G N E 4 JUNE4 24 _ _ _ _ _ ROCVILE MD.4~~44J FEBRUARY~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1973 .TMOso U.S. DEPARTMENT OF COMMERCE Frederick B. Dent, Secretary 4 ___~_____ NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION ____________ Robert M. White, Administrator NATURAL DISASTER SURVEY REPORT 73-1 Final Report of the Disaster Survey Team on the Events of Agnes A Report to the Administrator U. S. DEPARTMENT OF COMMERCE NOA; COASTAL SERVICES CENTER 2234 SOUTH HOESON AVENUE ,-.--. CHARLESTON, SC j 05-: Rockville, Md. February 1973 Property of CSC Library tK)~Fbury17 cv3 FOREWORD On June 23, 1972, during the widespread floods that accompanied Hurricane Agnes, a disaster survey team was designated by the Administrator of the National Oceanic and Atmospheric Administration to review the effectiveness of NOAA's storm and flood warning services and to gather de- tailed first-hand information from the communities within the river basins effected by the flood events of Agnes. The field survey was completed by June 30. Since then, several reports concerning the flood disaster have been published, the principal one being The Agnes Floods-A Post-Audit of the Effectiveness of the Storm and Flood Warnings System of the National Oceanic and Atmospheric Administration, A Report for the Administrator of NOAA by the National Ad- visory Committee on Oceans and Atmosphere, November 22, 1972, Washington, D.C. The post-audit report is supported by NOAA's self-exarnination and self-analysis in this "Final Report of the Disaster Survey Team on the Events of Agnes," which was made available October 1972 in prepublication form and is now published as NOAA Natural Disaster Survey Report 73-1 - C. E. Roache Deputy Associate Administrator for Environmental Monitoring and Prediction ii PREFACE On June 23, 1972, Robert M. White, Administrator of the Commerce Department's National Oceanic and Atmos- pheric Administration, designated a Disaster Survey Team to collect and report on the events pertaining to Agnes and the associated floods. The Disaster Survey Team included the following NOAA personnel: C. E. Roache, Team Leader Deputy Associate Administrator for Environmental Monitoring and Prediction E. J. Cartwright Meteorologist, Division of Meteorological Services, Headquarters, NOAA G. A. Baker Public Affairs Officer, Office of Public Affairs Headquarters, NOAA Ralph F. Kresge Assistant to Associate Director, Hydrology National Weather Service Harold A. Scott Chief, Public Weather Services National Weather Service John H. Thomas Regional Hydrologist, Headquarters Eastern Region National Weather Service Walter Seibert Chief, Weather Analysis and Prediction, Headquarters Eastern Region National Weather Service Gerald Shak User Services Representative Headquarters Eastern Region National Weather Service The Disaster Survey Team met initially in the office of the Team Leader on the morning of June 26, to review the task at hand and to develop a plan for the survey. The East- ern Region of the National Weather Service includes nearly all of the area involved. Therefore, the Team, at the request of the Director, Eastern Region, traveled that afternoon to the Region's headquarters at Garden City, N. Y., for a thor. Mi ough discussion by the Director and his staff of activities related to the flood disaster. The River Forecast Center (RFC) at Harrisburg, Pa., is responsible for a large portion of the area covered by the storm. The entire team moved to Harrisburg on the morning of June 27 to gather more detailed and first-hand informa- tion on the warning activities and on the reactions of State and local officials, the news media, and the public. At Harrisburg, the Team was divided into four groups of two men each. Each group was assigned specific areas to visit to obtain the facts on the warning services provided, adequacy of facilities, and public response. The groups com- pleted their initial fact-finding missions and returned to their home offices by Friday evening, June 30th. Major river basins and communities visited were: 1. James-Appomattox Richmond, Va. Lynchburg, Va. 2. Potomac Washington, D.C. 3. Schuylkill Norristown, Pa. Philadelphia, Pa. Reading, Pa. Pottstown, Pa. 4. Susquehanna Lewistown, Pa. Sunbury, Pa. Williamsport, Pa. York-Lancaster, Pa. Wilkes-Barre, Pa. Harrisburg, Pa. Elmira, N.Y. Corning, N.Y. Hornell, N.Y. Binghamton, N.Y. Covington/Mansfield, Pa. Lawrenceville, Pa. Painted Post, N.Y. 5. Genesee Wellsville, N.Y. Rochester, N.Y. 6. Upper Ohio-Allegheny/Monongahela Pittsburgh, Pa. Olean, N.Y. Salamanca, N.Y. Wheeling, W.Va. After the initial review by the Survey Team, task teams were organized to make more detailed studies of certain of the system's functional areas, to determine as accurately as possible the events that had taken place. 1V CONTENTS Foreword.................... Preface ..................... ii Executive Summary ................1 Chapter 1. The Hurricane Agnes Floods............. 3 Chapter 2. System Performance........... 5 Performance in Individual River Basins . ..... 5 Hurricane Forecasts. .........................26 Central Guidance Products ...........26 Data Collection. .............................27 Satellites. ..............................27 Radar..................28 River and Rainfall Reporting Networks. ......28 Flood and Flash Flood Forecasts- Watches and Warnings...........29 Dissemination of Forecasts-Watches and Warnings. 30 Public Response................31 Facilities and Staffing..............32 Appendixes: A. The Flood and Flash Flood Warning System ... 33 B. Winds During Hurricane Agnes........37 C. Precipitation During Hurricane Agnes.....45 V EXECUTIVE SUMMARY The Storm action by communities was responsive. Early pro- Though Hurricane Agnes, the first Atlantic hurri- jections of crest heights, which were low, were sub- cane of the 1972 season, was not an unusual storm sequently updated in response to continued rain. in the beginning, it eventually caused what has been Rapport between WSFO Washington and the Dis- termed the greatest natural disaster ever to befall trict of Columbia Civil Defense Operations Center this Nation. Formed from a depression off the coast requires improvement. Incidents of unnecessary loss of Yucatan on June 15, the storm developed and of life reflected a lack of public acceptance of the moved slowly northward, dumping large amounts seriousness of the situation. of rain on western Cuba and spawning tornadoes Schuylkill River Basin: Early forecasts of flood over the Florida peninsula and Keys. The sustained and flash flood conditions were timely and, although winds in Agnes never reached more than minimal low, brought good response from appropriate action hurricane intensity, but its circulation and precipi- agencies. After June 22, no river gages above Read- tation patterns covered extremely large areas. ing were operating; information from a gage at When it crossed the Florida coast near Panama Reading made possible accurate predictions for City on June 19, Agnes had degenerated to a tropical points downstream. Evacuation and rescue efforts storm. The storm then moved over Georgia and out prevented large losses of life. The principal problem into the Atlantic, up the coast to New York, and was dissemination of forecasts and warnings to the westward over New York and Pennsylvania (fig. 1). public. Along the way, Agnes regenerated in strength, pro- Genesee River Basin: The initial flash flood warn- duced excessive amounts of precipitation, and caused ing for the basin's headwaters followed the time of rivers and streams from the Carolinas to New York flooding in several communities. Subsequent issu- to rise to record or near-record stages. A record ances provided good lead time for protective action $3.5 billion in property damage was caused by floods downstream. No loss of life was reported on the and flash floods, and 118 persons were killed. Genesee, and understanding of the warnings was Evaluation exceptionally good. Good coordination between the The predicted movement of Hurricane Agnes as Corps of Engineers and the Rochester Weather it approached the Florida coast was excellent. While Service Office averted a potential dam failure. the winds were over-forecast, the problem was corn- Susquehanna River Basin: Flood and flash flood pounded by the news media which emphasized the warnings for the Susquehanna River Basin ranged higher gust figures as the predominant storm wind. from excellent at Wilkes-Barre-where the long lead This caused some adverse public reaction, but gen- time permitted evacuation of up to 100,000 persons erally public reaction and response were good. and prevented major loss of life-to warnings with Following are the Disaster Survey Team's evalu- minimum lead times, as at Harrisburg. Public dis- ations of the warning services provided: semination of warnings for small towns along the James River Basin (Including the Appomattox Chemung River was inadequate. Public response River): Both flash flood warnings and forecasts of varied from excellent to poor. The reason for the river crests were timely and allowed effective pro- great variation in performance of the warning sys- tective action. The response of local action groups tem was the erratic nature of torrential rains, which was positive. In some communities, dissemination of in some cases brought very rapid river rises. On information to the public was inadequate, because balance, according to Gov. Milton I. Shapp, the of the time-consuming nature of telephone dissem- Weather Service has "every reason to be proud" ination. of its performance. Potomac River Basin: Warnings of flash floods Upper Ohio (Allegheny-Monongahela) Basin: In and river crests were adequate. Generally, protective southwestern New York State, flash flood watches I 8:00 PM. JUNE 22 L _"_(DOUBLE CENTER) 8 00 AM. JUNE 23 800 AM JUNE 22 800 PM JUNE 21 JUNE 21 800 PM JUNE 20 NOTE ON JUNE 22, A SECOND LOW PRES- !~80O0 A.M SURE AREA FORMED IN THE STORM CEN- JUNE 20 TER THIS iS INDICATED BY THE DOUBLE 800 PM CENTER SHOWN FOR THAT DATE JUNE 19 were issued well in advance of flooding Local knowl- Figure 1.-Path of tropical storm Agnes, edge of flash flood characteristics of the river in this June 1972 Note On June 22 a second area resulted in prompt community action and low-pressure area formed in the storm timely evacuation of endangered areas Because center This is indicated by the double of the cooperation of county civil defense and local center for that date safety officials, there were no injuries and no loss of life. At Pittsburgh, flood warnings were generally avail- able with only minimum lead time, because of the suddenness of torrential rains and the fact that they occurred at night, at which time dissemination is most difficult. No lives were lost. 2 CHAPTER 1 The Hurricane Agnes Floods The great East Coast flood disaster of June 1972 western New York, releasing rains totaling 19 inches began with the landfall of Hurricane Agnes near in some areas. Panama City, Fla., on Monday, June 19. For one The maximum sustained winds over land were full week thereafter, the eastern seaboard was under 25 to 45 m.p.h. Because Agne's large circulation the continuous assault of torrential rains and floods brought an easterly-southeasterly flow over Florida, and flash floods which became progressively worse winds along the east coast were often as strong or as the week passed. Hardly a river basin north of stronger than those along the west coast, closer to Georgia and east of the Appalachians as far north the storm. Jacksonville, for example, recorded the as New York State was unaffected. Many of the highest wind gust in Florida, 56 m.p.h., early on basins experienced floods far exceeding the pre- June 19 when Agnes was heading for the panhandle. vious record, and near-record levels were reached Even at its peak, Agnes was a minimal hurricane. in others. This storm is unique in the annals of Over the open Gulf, maximum sustained surface natural disasters in the United States, in that it winds reached 85 m.p.h. on June 18, and surface caused disastrous floods and flash floods almost pressure fell to 978 mb on the 19th. Neither the simultaneously over such a large area. eye nor the wall cloud ever became fully developed. Agnes, the first Atlantic hurricane of the 1972 By the afternoon of the 18th, two things were ob- season, was not an unusual storm in the beginning. vious: Agnes would cross the coast along the Florida Formed from a depression off the coast of Yucatan panhandle; and the most destructive blow would be on June 15, the storm developed and moved slowly storm tides along the west coast. northward, dumping large amounts of rain on These tides hit the west coast on the morning of western Cuba and spawning tornadoes over the June 19. At Fort Myers, tides rose 3 feet above Florida peninsula and Keys. The winds in Agnes normal. A short time later, they were 4 to 5 feet never attained more than minimal hurricane in- above normal in the Tampa-St. Petersburg area. In tensity, but the area covered by the storm circula- the afternoon, Cedar Key recorded a tide 7 feet tion was exceptionally large. Its slow development above normal. As the storm neared the coast, Apa- and movement permitted a large amount of moisture lachicola recorded a tide 6.4 feet above normal. to be transported from the deep Tropics into the Agnes moved ashore as a tropical storm near Pan- storm system. This accounts for the extraordinarily ama City late in the afternoon of the 19th. Sus- heavy precipitation associated with the storm all the tained winds were 40 to 45 m.p.h., and gusts close way from Cuba through the eastern tier of States to the center reached 45 to 55 m.p.h. from Florida into New York. Precipitation produced by Agnes caused rivers After crossing the U.S. coastline, Agnes weakened and streams from the Carolinas to New York to to become a tropical storm and, by the time it turned rise to record or near-record stages. The most recent into Georgia on June 20, was classified as a tropical floods in this area approaching the same magnitude depression. On June 21, as the circulation moved were the 1936 Pennsylvania floods and the 1969 out over the Atlantic, next to the Carolinas, the Hurricane Camille floods in Virginia. In the wake storm began to regenerate and deepen. It moved of Agnes, previous record stages were exceeded by north along the coast to New York City on June as much as 12 feet on the Chemung River at Corn- 22, producing record rains all along the way. North ing, N.Y. (previous record 24.4 feet in 1946; Agnes, of New York City, Agnes joined forces with a large- 36 feet). On the Schuylkill River at Reading, Pa., scale circulation of colder air, and the combined the record flood of 1850 (26 feet) was exceeded by system turned westward over Pennsylvania and 5�/2 feet. Flood protection works, designed to pro- 3 vide protection against floods of the magnitude of and warnings were issued to the public by the Na- the previous record, were overtopped. River gaging tional Weather Service (NWS)-a component of the stations that were located to accommodate all rea- National Oceanic and Atmospheric Administration sonable stages were covered by water and washed (NOAA). Appendix A describes the flood and flash away. flood warning system and its various products- According to preliminary estimates, Agnes and the bulletins, forecasts, watches, and warnings. subsequent floods caused property damage of more Agnes was so large in areal extent and affected so than three billion dollars, and killed 118 persons in many communities for several days, that all sup- the United States. An unknown number of those porting statistics are not included in this report. killed were fully aware of the existence of the gen- Details of the storm, precipitation, and floods are eral flooding, nevertheless they waded in rain-swollen published in NOAA Technical Memorandum EDS streams, tried to drive through flooded areas, ignored NCC-1, Preliminary Climatic Data Report Hurricane warnings, or attempted to save others Agnes June 14-23, 1972, August 1972, issued by Consderng te sopeof te dsaser-wichthethe National Climatic Center, Environmental Data Consderng te sopeof te dsaser-wichthe Service; and in Preliminary Reports on Hurricanes President has called the greatest in the history of and Tropical Storms, Hurricane Agnes June 14-23, this country-loss of life was remarkably low. 1972, September 1972, issued by the Office of Me- Major flooding occurred in the James, Potomac, teorological Operations, National Weather Service, Schuylkill, Genesee, Susquehanna, and Upper Ohio which lists all bulletins issued by NOAA units dur- River basins. Numerous flood and flash flood watches ing the storm period. 4 CHAPTER 2 System Performance PERFORMANCE IN INDIVIDUAL RIVER BASINS burg. Local weather summaries and flash flood The major flooding-except for that on the upper watches and warnings are issued by WSFO Wash- Ohio River-occurred in the area served by the ington and WSOs Richmond and Lynchburg, when Harrisburg River Forecast Center. River District warranted. Offices most directly involved were: WSO Rich- WSO Richmond disseminated flood and flash flood mond for the James and Appomattox; WSFO Wash- warnings by means of local weather teletypewriter ington for the Potomac and Rappahannock; WSO loop and telephone to the Associated Press, United Trenton for the Schuylkill and Brandywine; RFC Press International, Richmond radio and television (RDO) Harrisburg for the Susquehanna in Penn- stations, and civil defense and city officials. The sylvania; WSO Binghamton for the Upper Susque- local teletypewriter loop also was used to dissem- hanna; and WSO Rochester for the Genesee. The inate civil defense and City of Richmond advisories upper Ohio, including the Allegheny and the Mo- to the news media. This procedure was very effective nongahela, is served by the RFC Cincinnati and the in informing the public. The staff at Richmond was WSFO (RDO) Pittsburgh. two people short during this emergency because of The floods generated by Agnes demanded the vacancies, but an additional meteorological techni- maximum participation by NWS staff throughout cian was detailed to Richmond for the period of the operational forecast system. All systems-includ- June 22 to 25. ing communications, data acquisition, data process- WSO Lynchburg is normally a part-time station, ing, and community-action programs-were strained but it went into 24-hour operation during the emer- and in some cases failed during this record catas- gency, starting Monday, June 19. An additional trophe. meteorologist was detailed to WSO Lynchburg from Performance of the system in each river basin WSFO Washington. To disseminate flood warnings, is described in the following summaries. Various the Lynchburg office must place telephone calls to operational problems associated with flood detection about 30 city, civil defense, and industry officials. and forecasting, and with alerting processes for com- The staff had great difficulty in completing these munities, are presented in the descriptions for key calls, and in some cases the issuance of warnings locations within each river basin. was delayed an hour or more. The first forecast for "showers and some locally James River Basin (Including the Appomattox heavy thunderstorms," for the night of June 19 and and Roanoke River Basins) all day on June 20, was issued by WSO Richmond Both flash flood warnings and forecasts of river at 5:45 p.m., June 18. Moderately heavy rains began crests were timely and allowed effective protective during the evening of June 19, and rain was heavy action. The response of local action groups was through the afternoon and evening of June 20. positive. In some communities, dissemination of in- The first flood warning bulletin for the James formation to the public was inadequate, because of River was issued by WSO Richmond at 10:00 a.m. the time-consuming nature of personal telephone on June 21. This bulletin, indicating the James River communication. would be 2 to 3 feet above flood stage at 7:00 p.m. on June 23, was a 57-hour prediction. It was Weather forecasts for the James and Appomattox changed at 11:15 a.m. the same day to reflect a River Basins are prepared at WSFO Washington. crest of 23 feet (11 feet above flood stage) on the WSO Richmond serves as the River District Office. 23d at 7:00 a.m.-serious flooding compared with River stage forecasts are prepared by RFC Harris- that caused by Hurricane Camille in 1969. These 5 JAMES RIVER AT HOLCOMBS ROCK, VA. Z 2.0- co1.5- STORM TOTAL BASIN PRECIPITATION =5 23 INCHES 1.0-- 0 40 , , ., 5o~ ..., ~ ~ ~ ."'-, RECORD STAGE 35 5 FT -1969 c'o I-. i~ I ' --- OBSERVED CREST 32 4 FT--221130 30- \ &AFLASH FLOOD WARNING I PREDICTED STAGE FLOOD STAGE-22\FT H---FORECAST TO STATED TIME ~20-- NO REPORTS AVAILABLE OPERATIONALLY 211500 TO 251200 I I I I I I I 0 20 21 22 23 24 25 26 27 DATE (JUNE 1972) bulletins were revised periodically through June 22, Civil defense, State, and city governments acted finally reflecting a crest of 28 feet at 3 00 a m. on quickly and in the best interests of protection of life June 23 The river crested at 28 6 feet on June 23 and property Public understanding of warnings was at 10 00 am. especially good, probably as a result of experience Of nine river gages in the James River, seven be- in the 1969 Camille floods. came inoperative between 7 00 p m and midnight on James River at Holcombs Rock, Va. June 21, well in advance of cresting Flooding in Holcombs Rock is located several miles upstream downtown Richmond knocked out longline teletype- from the city of Lynchburg and below the 3,250- writer, local teletypewriter, and facsimile circuits for square-mile drainage comprising the upper James a 16�/2-hour period beginning at 10 30 am on River Basin Rain began in the upper basin on the June 23 morning of Tuesday, June 20, and continued at a In the Appomattox River Basin, no previous moderate rate for 24 hours During the morning flood-stage base was available for determining flood of Wednesday, June 21, 2 8 inches fell in a 6-hour impact at Petersburg The one and only river gage period, adding to the 2 inches already on the ground ceased operating at 8'00 p m on June 22. The river began to rise rapidly By midnight, it had Virginia State Police reported five flood fatalities risen 22 feet in 24 hours and was within one-half along the James River, including one person who foot of the crest which occurred 12 hours later at had deliberately passed a police barricade. 11'30 a m on Thursday, June 22 The area had 6 JAMES RIVER AT RICHMOND, VA. (CITY LOCKS) 2.0 40 C- U~~~~~~.:, OBSERVED CREST-36.5 FT.-231600 C - I 8~~~~~~~~~ PREVIOUS RECORD 28.6 FT.-1969 W~~~~ 0 ' ~ ~~~~ ~~~~~~~~~~~~~~ - PREDICTED STAGE %~ ~----FORECAST TO S TATED TIME ,'i ~ ~ao-n GAG IOEAIEATR2100 0 LOOD TAGE- FT.NOTE: 23 PM: Afternoon of clay indicated N I' OF-~ ~ ~~~~~o 20 .A 2 22 2 ' 2 4 26 27 DATE (JUNE 1972) been placed under a flood watch at 6 a.m. on June Governor Holton of Virginia praised NWS in 21, and the first warning was issued at 9:20 a.m. his several radio and television appearances. that day. At this time, the stage was still 4 feet At Lynchburg, Central Virginia Industries, an below flood. association of manufacturers and industry, expressed Protective measures were only partially effective. concern about poor and late dissemination of fore- Some warning recipients took effective action, and casts and warnings. others did not. A plant operated by Libby-Owens- James River at Richmond, Va. (City Locks) Ford reported $1 million damage. Damage to com- The storm began slowly in the James Basin about mercial interests and permanent and mobile homes noon on Tuesday, June 20. After 24 hours of light- was fairly heavy. to-moderate rain, the intensity increased and in the The river gage survived the flood, but telephone last IS hours of Wednesday, June 21, the 6,757- lines were submerged at 3 p.m. on Wednesday, June square-mile basin received 4.6 inches, contributing 21. This was 20 hours before and 6 feet below the to the storm total of 6.6 inches. The heaviest crest. No readings were available during the next amounts were in the center of the basin near Scotts- 4 days. ville. The power loss to the RFC computer at Harris- Flood stage at Richmond is 9 feet. The first flood burg did not affect the preparation of forecasts for warning consisted of a forecast for an 11-foot crest this point. prepared and issued by Richmond RUG at I11 a.m. 7~ ROANOKE RIVER AT ROANOKE, VA. ~1.0- CK LO-STORM TOTAL BASIN PRECIPITATION=5 12 INCHES C -0~~~~~~11111 ~~~~~ - 0~ _ '~~-BEVDCRS-96F 220 I..- ~ P~wu R E O D2 F 4 /TELEMARK A FLASH FLOOD WARNING - / ~~~~OUT OFI ]SERVICE FLOOD STAGE , io T 2- 221200 ' ~ ~ TELEMARK RESTORVE ~~~10~~~~FLOO~~~DASTAE (JNI9 2 I I I ~ I I I.... ... 20 21~ 22 '23 24 -25 26 27 DATE (JUNE 1972) on Wednesday, June 21 Flood stage was reached 16 by the National Guard Four of the five bridges hours later, at 3 a m on Thursday, June 22 across the James River were closed During the The rise began at noon on Wednesday, June 21, rise, the water puiification plant was inundated, and and the level rose steadily for 54 hours, cresting at Virginia Electric Power Company lost one electric 4 p m on Friday, June 23 The level rose 36 feet power generation station and its dispatch center By and reached a maximum of 36 5 feet. This was 27 5 Friday, June 23, all of downtown Richmond had feet above flood stage and 8 feet above the previous ceased to operate, being without electric power, flood of record set in August 1969 drinking water, or communications Precautionary measures in Richmond were ex- The City Locks river gage was inundated but not tensive and extremely well executed An NWS severely damaged The telephone telemetering de- bulletin issued at 11 00 p m on Wednesday, June vice (telemark) ceased to function at 4 p m on June 21, and calling for a 28- to 29-foot stage resulted 22 After that time, a leveling party from the Rich- in the city and State emergency centers being mon Bureau of Survey made half-hourly readings, manned Warnings were disseminated through the relating bench marks to the water surface and pro- City Harbormaster's office Evacuations began on ducing an excellent stage record. the morning of Thursday, June 22, as the river- Roanoke River at Roanoke, Va. then slightly above flood stage-rose one-half foot Rain started in the basin on the afternoon of June per hour Residents of the low-lying areas were of- 19 and was fairly light until late in the evening of ficially served a legal order to vacate Closures were June 20 Within a period of about 10 hours, ending made in the dikes early that morning As the water at 7 a.m. on June 21, the basin received 3 2 inches rose higher, portions of the city were cordoned off of the 5 23-inch storm total An extremely sharp 8 river rise began during the night, and by daybreak Watches and warnings were issued by WSFO the 10-foot flood stage was reached with the level Washington to Associated Press, United Press Inter- increasing at the rate of 1 foot per hour. A flash national, local radio and television, Red Cross, and flood watch was issued by Washington WSFO at civil defense, by means of a local teletypewriter loop 5:30 a.m. on June 21 and a warning 45 minutes and VHF-FM radio. Other offices were notified by later. The level rose rapidly through the day, crest- telephone. The office's telephone warning list re- ing at 9 p.m. at a stage of 19.6 feet. This is 9.6 feet quires 44 calls. above flood and exceeds the previous record set in A flash flood watch was issued for northern Vir- 1940 by 1.4 feet. ginia, to include the counties immediately west of The actual forecast operation, conducted by WSO Washington, D.C., at 6:00 p.m. on June 20. A fore- Roanoke and RDO Raleigh, N.C., consisted of cast for "heavy rain at times" was issued at 9:40 issuing warnings and advisories of a descriptive a.m., June 21. This forecast was for the immediate nature. The only statements that might be considered forecast period (today). When heavy rain began actual stage forecasts were those issued at 11:25 during the morning, the flash flood watch was a.m. and 6:00 p.m. on June 21. The former stated changed to a warning at 12:45 p.m., June 21. that this flood would approach the record of 18.25 A flash flood warning for the Washington area was feet. At the time of issuance, however, the stage was issued Wednesday, June 21, at 4:45 p.m. At 6:13 already 15.5 feet, and the river was rising 1 foot p.m., based on a radar report, the civil defense and per hour. The 6 p.m. statement declared that the police of Alexandria were notified by telephone of river was near crest, and it was within 0.6 feet and impending heavy rain, and evacuation of Four-Mile 3 hours of it. The foregoing is not to be construed as Run was recommended. A few hours later, Four- a criticism of the operation. This RDO is not served Mile Run had risen to record flood level. At 6:00 by an RFC and must prepare its own forecasts by p.m. on June 21, WSFO Washington, in its capacity means of a rudimentary procedure. Cooperation by as a Hurricane Warning Office, issued a bulletin on radio and TV stations is said to have been excellent, tropical storm Agnes, indicating that large stream As a result of their broadcasts-and of the efforts flooding was expected to be near record-high level of the local civil defense organization, and a limited throughout the Carolinas and Virginia that night and number of calls by NWS personnel-the warning farther northeast Thursday. was spread very quickly. It was received with apathy Initial flood crest forecasts were low. However, by some people who did not believe the water could subsequent predictions, reflecting the continuous rise high enough to bother them. Others, however, heavy rains, repeatedly raised the crest value until took action to reduce property damage. the final crest was accurately predicted for Little The river gage is equipped with a telemark con- Falls 11 hours in advance, and for Frederick, Md., nected through the switchboard of the Appalachian about 23 hours in advance. Predictions for the Wis- Power Co. This went out of service at 11 a.m. on consin Avenue gage, which is the most important June 21, when the connecting lines were inundated. gage for Washington, were complicated by the loss Service was restored 25 hours later. In the interim, of readings from the gaging site, substitution of power company employees supplied slope-gage read- readings from a previously unused gage at Key ings. Bridge, and the loss of communications with RFC Harrisburg. Potomac River Basin The river forecasts for the Potomac and Monocacy Warnings of flash floods and river crests were ade- Rivers are normally prepared by RFC Harrisburg quate. Generally, protective action by communities and disseminated by WSFO Washington. Because of was timely and responsive. Early projections of crest communications outages between the two offices, heights, which were low, were subsequently updated WSFO Washington prepared the later forecasts. in response to continued rain. Communications be- There was not optimum coordination between the tweenWSFOWashngto andthe istrct o Co-There was not optimum coordination between the tween WSFO Washington and the District of Co- WSFO Washington and the District of Columbia lumbia Civil Defense Operations Center require im-W ahntnadteDsri t o C l u ba- ~lumbia Civil Defense Operations Center require im- Civil Defense Operations Center. The Center, which provement. Incidents of unnecessary loss of life re- includes the mayor's emergency command post, did flected a lack of public acceptance of the seriousness not have the information needed to equate crest not have the information needed to equate crest of the situation. heights with the potential flooding impact. Weather forecasts and flash flood watches and The public's primary source of information was warnings for the Potomac River Basin are prepared radio and television. Surrounding communities re- by RFC Harrisburg and disseminated through WSFO acted well, but there were some minor problems. (RDO) Washington. For the most part, public understanding and reaction 9 POTOMAC RIVER AT LITTLE FALLS NEAR WASHINGTON, D.C. 6 4PRECIP AT DULLES AIRPORT 5- rSTORM TOTAL=13 6 INCHES 4- MEAN PRECIPITATION FOR _ BASIN ABOVE LITTLE FALLS STORM TOTAL=7 5 INCHES 2- 0 � FLASH FLOOD WARNING I:1- - �0 PREDICTED STAGE xo *i i.E I I-,-- FORECAST TO STATED TIME 0 0' o a V) 0 400 Q I 400 <ID c---OBSERVED CREST-22 0 FT -240200 =020 .g ~, oo~-, o-' / 7 , \FLOOD OF RECORD-484,000 CFS 300- -18 3 00 / \ ABOUT 25 FT -1936 18 4 04-16. �8 ~ ~ ~ ~ ~~~0 V /DISCHARGE AT LITTLE FALLS t 2 200- -14 FLOOD STAGE-12 FT --12 tl>DISCHARGE CONTRIBUTION OF ff8100-' LOCAL AREA BELOW POINT OF -10 100-- . * ROCKS AND FREDERICK 9% OF BASIN ' 14% OF RUNOFF VOLUME - ', .- ' - I I I -4 20 21 22 23 24 25 26 27 DATE (JUNE 1972) was slow, perhaps bordering on disbelief Two lives Wednesday, June 21. it became heavy in the basin were lost in the District of Columbia when a family with the greatest concentration in the local area went wading in Rock Creek and the two children immediately above Little Falls Dulles Airport, which were swept away from their parents The Red Cross is In this area, recorded 5 74 inches in one 6-hour reports 21 lives lost at unspecified locations In Mary- perod and 11 88 inches in the 24-hour period ending land. at 7 a m on Thursday, June 22 The heavy rains caused an almost immediate rise of the river Flood Potomac River at Little Falls Near Washington, D.C. stage of 12 feet, caused solely by rainfall in the I ittle Falls is the last station on the main stem of local area, was reached 10 hours after the rise began the Potomac above tidewater Stages at this gage An initial peak of 13 5 feet occurred at 2 p m on do not directly relate to flood problems in the Wash- Thursday, June 22 This was followed by a slight ington, D C, metropolitan area, but discharge fore- drop as the local area runoff receded. Then, as the casts for Little Falls are used to produce the stage water from the main portion of the basin moved forecasts for the Wisconsin Avenue gage located in in, a secondary rise began and continued for 32 the tidal reach hours The crest of 22 feet was reached on Satur- Rain began in the basin about noon on Tuesday, day, June 24, at 2 a m This was 10 feet above flood June 20, and continued at a light-to-moderate rate stage but about 3 feet below the record flood of for 24 hours During the afternoon and evening of March 1936. 10 POTOMAC RIVER iAT WISCONSIN AVE., WASHINGTON, D.C. STORM TOTAL BASIN PRECIPITATION = 7.51 INCHES CO RECORD FLOOD-17.7 FT.. 1942 ~ . OBSERVED PEAK 1 5.4 FT 240900 :FLASH FLOOD WARNING I -- 'C-- o / �PREDICTED STAGE 21 22 2 20 00 0 27 OIi��� 1 -- F RECORD TLO STATED JTU flas floo warningw i-s :4 0 -0 .t in in- FLO O STAGE- F "' I I :2 " : : : 2~ i : : 26 ' - 27 : I DATE: \J-FRANE 1972)T S TM watch since 9:25 p.m. on Tuesday, June 20, and a rain in the local area. The reporting network has fash flood warning was issued at 4:45 p.m. on Wed- very few gages in that area, and the intense rains nesday, June 21, at about the time the heavy rain were not adequately sampled. began. The continuing rain required frequent up- While observed rainfall is the quantity used to dating of stage forecasts issued during the rise. forecast the initial response at this gaging station, There is not much property subject to flooding in prediction of the main flood wave is based primarily the vicinity of the Little Falls gage. Those persons on observed discharge at upstream points. The final on the warning list were notified, and the forecasts forecast for Little Falls is a function of the observed were further publicized through civil defense and hydrographs at Point of Rocks on the main stem local radio and TV stations. Because of the short and Frederick on the Monocacy River, the prin- lead time, precautions were minimal and damage cipal tributary to the lower Potomac. RDO lost con- was heavy. Numerous homes in the Seneca area tact with the river-gaging stations at Frederick and were badly damaged, as were recreational facilities Point of Rocks 37 hours and 19 hours, respectively, along the river. before the crest was reached at Little Falls. The forecast operation was extremely difficult both Another complicating factor was the fact that, for RFC Harrisburg and for RDO Washington. A while this was no t a record flood for this reach of number of factors c ontributed to th is. The initial the Potomac, it was the highest since the gaging and 11 forecast point had been moved to the present site was reading 1 4 feet higher than the gage at Wis- in 1965 Consequently, the stage-discharge relation consin Avenue for Little Falls was an extension above the previous NWS was criticized-not because of the quality maximum experienced at this site. This extension of the forecasts-but because those forecasts con- has been found to be in error by 1 8 feet. sisted only of anticipated stages The critics main- The Little Falls gage remained operative through- tained that NWS personnel should have advised them out the event what land areas would be inundated and what action The forecast operation was affected by loss of should be taken power to the RFC computer at 8'00 a m on Friday, While the flood in the Washington area was not June 23 Final forecasts had to be prepared man- disastrous, it caused fairly heavy damage to both ually private and public property Four deaths were re- Potomac River at Wisconsin Avenue, Washington, ported in the immediate area D.C. Because of the loss of power to the RFC computer This gage, located at the foot of Wisconsin Av- at Harrisburg, the last two forecasts for this point enue in Georgetown, is the principal forecast point were based on manual computations. for the tidal reach of the river at Washington. The Schuylkill River Basin response of the river at this point to rainfall in the Early forecasts of flood and flash flood conditions basin occurs about 2 hours later than at Little were timely and, although low, brought good re- Falls. Consequently, when the intense storm oc- sponse from appropriate action agencies After June curred on the evening of Wednesday, June 21, the 22 rivet gages above Reading were not operating, level rose sharply, and passed the 7-foot flood stage observations from a gage in Reading made posible at midnight An initial peak of just over 10 feet accurate predictions for points downstream Evacu- occurred late in the morning of Thursday, June 22 atilon and rescue efforts prevented large losses of This was followed by a brief fall Then, as the main life The principal problem was dissemination of flood wave moved in, the level again began to rise forecasts and warnings to the public to a crest of 15.4 feet at 9 a m. on Saturday, June Weather forecasts for the Schuylkill River Basin 24. This was 8 feet above flood stage, but 2 3 feet are prepared at WSFO Philadelphia. River stage below the record flood of 1942. forecasts are normally prepared by RFC Harrisburg A flash flood warning was issued at 4:45 p m. on However, during the widespread flooding, WSO Wednesday, June 21 Normally, a warning of this Trenton, using previous RFC guidance and the sys- type would be considered applicable to small streams tem's forecast procedure, issued river forecasts for in the area but not to lands adjacent to the tidal the Schuylkill River Area flash flood watches and reach of the Potomac. There was, in fact, no indi- warnings are also issued by WSFO Philadelphia and cation at this time that the stage at Wisconsin Av- by WSO Trenton Action agencies received the in- enue would rise above flood level within a few formation through State civil defense systems and/or hours The first actual stage forecast was issued State Police teletypewriter circuit. The general public Thursday, June 22, at 8 a m at the time the initial received warnings through radio and television, which 10-foot peak occurred It called for a continued rise were serviced by Associated Press or United Press to 14 feet later in the day When the level began to International (In this area, there are very few sub- drop a short time later, this was revised to 10 to 11 scribers to the NOAA Weather Wire Service.) The feet When the secondary rise began, forecasts were news wire services gave bulletin status to all weather steadily increased, and shortly after noon on Friday, warnings However, from 30 minutes to an hour June 23, called for a crest of 18 to 19 feet, which sometimes elapsed between the issuance of a warn- would have been an all-time record had it occurred. ing and its receipt by the radio/television studio Many aspects of this forecast and warning opera- The first forecast of heavy rain was issued by tion must be considered unsatisfactory All of the WSFO Philadelphia at 5'00 pm on June 20, mindi- technical problems that complicated the Little Falls cating heavy rain that night A flash flood statement forecast also affected the forecast for Wisconsin was issued by WSO Trenton at 7:30 a m., June 22, Avenue In addition, the telemetering device at Wis- for the upper Schuylkill. consin Avenue failed at 7 p m on Wednesday, June All river gages in the Schuylkill basin except the 21, when the rise had barely begun RDO Washing- power company staff gage at Reading became in- ton was able to obtain stage reports from a city operative during the afternoon of June 22. The employee who was reading a staff gage at Key result was a total lack of information on river stages Bridge, 1/2-mile upstream from the Wisconsin Av- above Reading, but forecasts for points below Read- enue gage At the time of the peak, the staff gage ing were good. 12 SCHUYKILL RIVER AT PHILADELPHIA, PA. C4 STORM TOTAL BASIN PRECIPITATION=8.26 INCHES , c='p~. ':': , il,.FLASH FLOOD WARNING - . ~<~ � ~~~~~~~~PREDICTED STAGE , ~ ~ ~ ~ ~ ~ ~ 4 5 I )BSERVEDCR EST-" FLOOD STAGE-11 FTT NOTE: 22 PM: Afternoon of day indicaoted____ - --- 23 AM. Morning of day indicated 23 PM: Afternoon of clay indicated NO REPORTS AVAILABLE OPERATIONALLY AFTER 230900 ~l ~~~~~ I I- 4FT 23090 ' C 2i.2.:~' ': :"OTE: '~ ' 22 ......M: ........oo:n of 25 26 27 DATE (JUNE 1972) NWS radars closest to the Schuylkill River Basin issuance of watches, which the public comes to are located at Patuxent River, Md., Atlantic City, ignore. N.J., and New York, N.Y. These radars do not pro- Schuylkill River at Philadelphia, Pa. vide the adequate information on rates of precipita- Rain began in the basin on the morning of tion over the basin that is needed for flood and flash Wednesday, June 21. The intensity was light during flood warning services. the afternoon and early evening, but became heavy An additional man was temporarily detailed by during the night. The area had been under a flash Eastern Region Headquarters to the WSO staff at flood watch since 4:30 p.m. on Tuesday, the 20th, Trenton, N.J., on June 21. The Meteorologist-in- and the first warning was issued at 5:30 p.m., Charge, WSFO Philadelphia, was not on duty, but Wednesday, in anticipation of the heavy rain. By the office had sufficient personnel available during daybreak on Thursday the 22d, the river was begin- the emergency. ning to rise and at 5:30 p.m. the 11-foot flood stage Although some action agencies were unable to re- was reached. The rain began to taper off during the late the river stage forecasts to probable flooding, afternoon. The river continued to rise through the civil defense and other action agencies responded night, cresting at 9 a.m., Friday, June 23. The extremely well and can be credited with keeping loss maximum stage was 14.7 feet, equal to the flood in of life to a minimum. Two lives were lost. Gen- 1933 but 3 feet below the record of 17 feet set in erally, people apparently did not appreciate the October 1869. severity of the flood. NWS received compliments Crest forecasts were issued throughout the period from the mayor, borough manager, and newspaper of the rise. Those issued shortly before and at the editor from Pottstown. time of the crest called for a secondary rise, 3 to 4 Newsmen in the area felt that the lack of public feet higher than the first crest. A secondary crest response may have been due to a too frequent did occur, but it was lower than the first. 13 Extensive precautions were taken in the Phila- The first forecast of rain and thunderstorms was delphia area, consisting of the placing of highway issued by WSFO Buffalo at 5:00 am, June 20. At barricades and the evacuation of residences in the 6.20 a m, June 21, WSO Rochester called Buffalo northern suburbs While damage was heavy in the to report that Scio, N.Y., had received 2 76 inches Schuylkill basin as a whole, it was light in and of rainfall in 12 hours, and the river was 2 feet above around Philadelphia. Three deaths were reported. flood stage. Ten minutes later, forecasters in Buffalo While the river gage survived the flood, the heard on commercial radio that schools m Wells- telemark was out of service from the time of the ville, N.Y., were closed because of flooding. These crest to the end of the event. During that period, two communities are located on the headwaters of stage reports were not available operationally. Fore- the Genesee A flash flood warning for the Genesee casts for this point are based, to a large extent, on River and its tributaries in the Scio/Wellsville vicin- stage readings from upstream points These were Ity was issued by the Buffalo forecaster and called almost completely lacking The stations (proceeding to Rochester for distribution at 7:00 a m., Wednes- downstream) and their reporting performance during day. The zone forecast was revised to include a flash the flood are as follows: flood warning. At 7:35 a.m., June 21, WSO Rochester reported Berne: Out of service permanently at 1 p.m on that Sclo had received an additional 1 6 inches of June 22. rain in 1 hour Flooding in Wellsville and Bolivar Reading: Reported crest of 31.5 feet at 4:30 a.m. areas, heavy rain m Steuben County, and overflowing on June 23. Report was received at RFC at 9.00 In Hornell, N.Y., were reported over NAWAS* at a.m. This crest was 9.5 feet above the previous 7 40 am The flash flood warning was extended to record of 22 feet set in May 1942. include Allegheny, Livingston, Steuben, Schuyler, Pottstown: Last report was at midnight on June Yates, and Ontario Counties and distributed by WSO 22. River stage was 20 9 feet and said to be rising. Rochester at 8:30 a.m that day. At 10.30 a.m., the city of Wellsville lost power, and communications It eventually went 9 feet higher The previous record of 21 feet was set in February 1902. were disrupted NAWAS was used by WSFO Buffalo, but time lags of up to an hour were caused by the Norristown: This is the last main stem station before need to read the warning bulletins slowly to New Philadelphia Therefore it is most important m mak- York State NAWAS headquarters in Albany before ing river forecasts for Philadelphia. While no reports the relay to Allegheny County could be effected. were received from this station at the time, the river The river forecast prepared by WSFO Buffalo at Norristown crested at 24.5 feet on June 23. The and disseminated by WSO Rochester is given m previous record of 21 feet was set in August 1933. table 1. Genesee River Basin The lower Genesee River Basm was subjected to The initial flash flood warning for the basin's head- the successive rains of two large weather systems on waters was isvued after flooding had occurred in June 21 and 22, which produced two separate but several communities Subsequent issuances provided cumulative flash floods in the Genessee headwaters good lead time for protective action downstream. on these days. There was no loss of life reported on the Genesee, Flooding along the main stem of the Genesee and understanding of the warnings was exceptionally did not take place until Friday, June 23. River state- good A potential dam failure was averted by good ments and warnings were Issued well in advance of coordination between the Corps of Engineers and flood occurrence. On June 24, the flood control dam WSO Rochester. at Mt. Morris, N.Y.-which is normally empty-be- WSFO Buffalo prepares zone weather forecasts came filled to capacity and threatened to cause a for upper New York State, including the Genesee major disaster if it were to collapse. The Corps of River Basin This basin is outside the jurisdiction of Engineers contacted WSO Rochester to advise them of the need to relieve pressure on the structure. Per- a River Forecast Center. River stage forecasts are prepared by RDO Rochester. Flash flood watches sonnel at RFC Hartford responded to WSO Ro- and warnings for the Genesee are issued through chester's request to calculate a safe flow level and WSFO Buffalo and transmitted to WSO Rochester advised that a flow level not to exceed 15,000 cubic by telephone and RAWARC* for distribution. Dis- feet per second would be required to prevent dis- semination is made through news wire services, astrous floodig downstream. Residents who would local teletypewriter loop, VHF-FM, and a telephone be affected by the flow were given 5 hours' notice warning list. There is no NOAA Weather Wire (by the Corps) to evacuate before the water was Service In New York State. *NAWAS-Natlonal Warning System, primarily a conference tele- phone system operated by Civil Defense 14 Table 1.-River forecast for Genesee River Date & time (EDT) Previous River Stage Forecast riverfirst Crest, Flood flood Date & time Forecast Effective date reached date & time Station stage record issued (EDT) stage & time (EDT) flood stage (EDT) GENESEE RIVER 17' 28.9' 21/10:45 p.m. 21.0' Unknown 22.0' Portageville 1956 23/12:00 noon released from the dam. Some flooding occurred as at 3:00 a.m. on Friday, June 23. This forecast stated a result of the relief flow, but the greater threat was that the Susquehanna at Wilkes-Barre was expected averted. to crest at 40 feet, at 8:00 a.m. on June 24, 7 feet No warnings were issued for the flooding that above the flood of record. The forecast triggered a occurred at Wellsville about 2:30 a.m., June 21. The mass evacuation of 80,000 to 100,000 persons by local radio station was critical of NWS. The station's civil defense authorities, and is unquestionably re- telephone call to WSFO Buffalo was answered by a sponsible for preventing a disaster of unimaginable recorded message of general weather information. magnitude. The station is on WSO Rochester's call list, but its Forecasts were distributed to the public and public staff asserts that no call was received. safety officials by means of the NOAA Weather The emergency situation in WSO Rochester was Wire, the news wire services, and available civil complicated by the retirement of the Meteorologist- defense and State Environmental Protection Agency in-Charge, who was absent on terminal leave. The systems. Personal telephone contact provided the slack was taken up effectively by WSFO Buffalo, only warnings for many areas. In Pennsylvania, the and the vacant position was filled quickly by dis- NOAA Weather Wire did not have sufficient sub- patching an acting Meteorologist-in-Charge from scribers to make it an adequate warning medium. In Albany. New York State, where the NOAA Weather Wire is not installed, broadcasters monitored other sta- Susquehanna River Basin tions and maintained contact by telephone. Normal Flood and flash flood warnings for the Susquehanna dissemination of forecasts through RDO Bingham- River Basin ranged from excellent-as in Wilkes- ton for the Chemung River is by telephone. Barre, where a long lead time permitted the evacu- Flood warning bulletin #1 was issued by WSO ation of up to 100,000 persons and prevented a (RDO) Binghamton, N.Y., at 9:30 a.m. on June major loss of life-to warnings with a minimum 21, for Steuben, Chenango, and Tioga Counties, lead time, as in Harrisburg. Public dissemination of which encompass the cities of Hornell, Painted Post, warnings for many small towns along the Chemung Coring, and Elmira. This bulletin, advising "all River was inadequate. Public response varied from excellent to poor. The reason for the great variation ditey," was distributed by telephone to a calling diately," was distributed by telephone to a calling list which included Elmira and Corning radio and erratic nature of torrential rains, which in some TV stations. At 10:15 am., the flood warning bul- cases brought very rapid river rises. On balance, letin was extended to include additional counties, according to Gov. Milton J. Shapp, NWS has "every and, at 11:00 am., the extended warning was tele- reason to be proud" of its performance. phoned to area radio and TV stations as part of Weather forecast responsibility for the Susque- the local forecast. At 4:00 p.m., June 21, flash flood hanna Basin is divided between the Pittsburgh, Phil- warning bulletin #2 was issued for Steuben, Che- adelphia, Albany, and Buffalo WSFOs. All river mung, Tioga, Broome, Chenango, Cortland, and forecasts are prepared by RFC Harrisburg. Otsego Counties. Although the bulletins were re- A flash flood watch was issued at 11:00 a.m. on peatedly broadcast, there was a consensus that "there the morning of June 21 by RFC (RDO) Harris- was no recollection of warnings." burg for much of the Susquenhanna Basin. A flash Crest forecasts issued during the progress of the flood warning issued at 3:00 p.m. that day, as storm required frequent upward revisions to reflect extremely heavy precipitation began, assured excel- the continuous heavy rainfall. This situation was lent lead time before critical stages could develop prevalent throughout the river basin. The perform- on smaller streams. ance of the local radio and television stations was Perhaps the most outstanding issuance of the exemplary. In general, they were the principal means whole disaster was a flood forecast sent by RFC of warning citizens and, in some cases, they were Harrisburg to the civil defense office in Wilkes-Barre the only avenue of warning, remaining on the air for 15 80 to 90 hours until the emergency was over. Most coming County Commissioner stations exercised good judgment In filtering rumors Both flash floods and river floods occurred at and avoiding sensationalism, refusing to broadcast Harrisburg, and the public was confused by the two reports of broken dams. for example, until the in- types of warnings. Flash flood warnings requiring formation could be verified Many small stations Immediate action were in effect while river flood invested large sums of money in long-distance tele- warnings were predicting flooding some hours later. phone calls to obtain continuous information on For example, the publisher of the Harrisburg Patriot weather and flood conditions News reported that he and his staff came to work The staffs of all NWS offices in the basin per- at 4 00 am. on Thursday morning, June 22. At formed admirably under extremely hazardous emer- 7 00 a m, the river stage on the Susquehanna at gency working conditions The staffing capacity for Harrisburg was 11.2 feet, 5 8 feet below flood stage. RFC Harrisburg was stretched near the breaking But at 10 00 a m. the newspaper staff was evacuated point as the effects of Agnes spread throughout the from its offices, with the loss of one life Flash flood entire assigned forecast area. warnings were in effect at that time because of heavy As the flood rose to above-record proportions, rains during the night The flooding of the Patriot 22 river gaging stations out of 60 on the river be- News was from flash flooding on Paxton Creek The came inoperative or were destroyed, and communi- Susquehanna reached the flood stage of 17 feet at cations systems began deteriorating Portions of the Harrisburg at 2 00 p m that evening. Federal-State Radio River and Rainfall Reporting The effectiveness of the support provided is de- Network became inoperative. scribed in the following letter from Milton J Shapp, Power failures were prevalent throughout the Governor of Pennsylvania, to the Hydrologist-in- Susquenhanna Basin, affecting the timely collection Charge of RFC Harrisburg' of rainfall and river information as well as the prep- "On behalf of the people of Pennsylvania aration and dlssemlnation of flood forecasts Failure and the Commonwealth Government, I ex- of the power system supplying RFC Harrisburg, at tend to you and your fellow workers m the 7 14 a m on June 23. was most critical The Center's Federal-State River Forecasting Service our staff then had to perform forecast operations man- sincere thanks for the highly valued service ually, under lantern light When the telephone and so capably rendered in connection with last teletypewriter system at RFC Harrisburg failed, month's disastrous flood time-concumlng emergency methods were employed "The June 20 flash flood watch, changed to collect substation reports and disseminate fore- to a warning the following day, provided casts The staff was able to make outgoing calls, people In the Lower Susquehanna River and this made it possible to obtain minimum data. Valley with initial notice and forecast of There were no signals to indicate incoming calls things to come During the next several Public understanding and reaction varied In the days, the widespread disruption of com- Covlngton/Mansfield area of Pennsylvania, one man municatlons systems and facilities admit- was awakened in his mobile home and informed of tedly resulted in some delays in the normal the flood approach He responded by turning over dissemination of crest forecasts and similar and going back to sleep Instances were cited of information Nonetheless, the vital mes- people refusing to leave their homes, necessitating sages did 'get through' dangerous rescue efforts later, which cost the life 'Should there be any doubt as to the of at least one rescuer in Painted Post, N Y At the value of the forecast operations, I need other end of the scale, many people in Selinsgrove, cite only the 40-foot crest predictions for Pa, reacted quickly enough to save personal belong- Wilkes-Barre, which came early Friday inrags and even appliances There was both criticism morning. June 23 Passed to the Luzerne and praise for NWS operations, leavened by a wide- County officials by our State Civil Defense spread feeling that "everyone was having his prob- Director, with a recommendation that lems" in this short-fuse phenomenon everyone 'behind the dikes' be evacuated, The Williamsport Mayor, John R Coder, the local that single bit of essential information un- radio and TV media, and the Lycoming County Civil questionably was responsible for the sav- Defense Director expressed praise for the dedicated ing of countless human lives, which other- service renderd by the staff of WSO Wlllamsport. wise would have been lost. In Wlliamsport, Pa, th malfunctioning of a river "You and your associates have every gage, used by WSO Wllllamsport for its reports to reason to be proud of your contribution the community, occasioned a complaint by the Ly- to our common safety" 16 CHEMUING RIVER AT CHEMNG, N.Y. STORM TOTAL BASIN PRECIPITATION = 8.70 INCHES ~-E S , ~E gg_.. OBSERVED CREST-31.4 FT. -240300 ~- 1o . PREVIOUS RECORD , 24 FT., 1946 0. 24 FT~, ,4 GAGE INOPERATIVE OR DESTROYED ,04~~~ ~ 230100 TO 251600 o-;I:; i a0 o _ 30-.... . : 30 FLOOD STAGE-12 FT. : Significant rainfall began in the basin on the eve- hampered protective and rescue operations. Infor- ning of June 20. The area had not been under a mation on the extent of these operations is sketchy. wa tch, and the firs t flash flood warning was issued Property damage in the area was extremely heavy. by RDO Binghamton at 8:30 a.m. on June 21. At While no deaths have been reported in Chemung, this time, sizeable rises had started in the head- there are said to have been 19 drownings in the waters but no rise had yet taken place at the station. upper basin near Corning. The rainfall through June 21 was associated with Susquehanna River at Wilkes-Barre, Pa. a cold front. By t h e time precipitation ended late On the morning of June 21 , no rain was falling on June 23, tropical storm Agnes had contributed in or near Wilkes-Barre, but the flood crest then about 4 inches for a 4-day total of 8.7 inches over developing above Chemung would eventually affect the basin. The heaviest amounts were in the upper this area. A flash flood watch was issued at this portion of the drainage. A crest of 31.4 feet oc- time and replaced by a warning during the after- curred on the morning of June 24. This is 19 feet noon. Rain began in the area immediately north of above flood stage and 7 feet above the previous Wilkes-Barre later on June 21 and continued through record of 24 feet in May 1946. The gage was de- June 2 3, producing 6 inches in the local drainage was 8 feet below the maximum. Final forecasts for duced a crest of 40.6 feet on the evening of June 24. this point were prepared manually after power to This is 18.6 feet above flood stage and 7.5 feet the RFC computer wa s lost. above the previou s reco rd set in 1936. P rep aration 17 SUSQUEHANNA RIVER AT WILKES BARRE, PA. @2o m 0. STORM TOTAL BASIN PRECIPITATION=589 INCHES I 0- rES~~~~~~~~C o. ~ ~ ~ 40-~ ~ ~ 3-0 0-- Or) CPRC T ED ST-A ~~~~~~ ~~~~.~ tJ Cr) C /) C'J 40- ~ g~g - -GA---OBSERVED CREST-40S6 FT -241900 3o- c X LL. L&~~~ ~ CC) CO 204 214 22C34 2 5 C4 FLOOD STAGE-22 FT I--- ----0 ~20-- 0 �AFLASH FLOOD WARNING'~ C cC c~ * PREDICTED STAGE C, -- ----- FORECAST TO STATED TIME %% D GAGE INACCESSIBLE AFTER 231900 I I I I I I I 20 21 22 23 24 25 26 27 DATE (JUNE 1972) for the flood on the basis of the initial warnings clock. Despite their efforts, the dikes on both banks and later forecasts was unusually good and involved were overtopped and eventually breached The the evacuation of 100,000 people Property damage breaching of the dikes probably resulted in a slight was heavy, but there is overwhelming evidence that decrease in the stage in the main channel and a a much greater loss was averted One death by considerable increase in the size of inundated area. drowning has been reported in the Wilkes-Barre area. West Branch Susquehanna River at Williamsport, Pa. Although the river gage was not destroyed, radio In the river basin above Williamsport, rain began communication with it was lost early in the water's slowly on the afternoon of June 20. On the morning rise. Later, inaccessibility of the gage required the of June 21, the intensity was increasing and the use of a wire-weight gage and telephone communi- area was placed under a flood watch. The first warn- cation with the local observer A forecast of a 40- ing was issued at 4 00 p m The rate of rainfall con- foot crest was prepared prior to the loss of power tinued to increase through the night and most of the to the RFC computer Although minor revisions next day. By early morning of June 25, 13 5 inches were prepared manually later, the decision to evac- had been observed at Williamsport Average pre- uate areas in back of the dikes was based on the cipitation over the entire 5,682-square-mile basin 40-foot forecast above Williamsport was 8.8 inches This heavy down- Several thousand teenagers, among others, volun- stream concentraton caused disastrous flooding along teered for sandbag duty and worked around the small streams in the local area The West Branch 18 W. BR. SUSQUEHANNA RIVER AT WILLIAMSPORT, PA. I5 - 1;5~ STORM TOTAL BASIN PRECIPITATION = 8.79 INCHES o_ ~ ~ ~ ~ L 0 40 _ - - .--OBSERVEO CREST-34.8 FT,-232130 __ - PREVIOUS RECORD-33.6 FT., 1936 30- 20-- C , ~ ~ ~ FLOOD STAGE-20 FT. :0- _A FLASH FLOOD WARNING * PREDICTED STAGE I--- FORECAST TO STATED TIME 0 2G ~~21 22 23 24 25 26 27 DATE (JUNE 1972) crest occurred at 9:30 p.m. on June 23 at a stage cipal crest forecasts were prepared before power to of 34.8 feet, 15 feet above flood and one foot above the RFC computer was interrupted. the previous record in 1936. The crest at Williamsport was only one foot lower The 34.8-foot stage is a wire weight gage reading. than the top of the dikes. The maximum stage recorded in the well house was Susquehanna River at Sunbury, Pa. 31.4 feet. Information available at this time indi- Sunbury is located on the main river just below cates that the discrepancy probably was the result the junction of the North Branch with the West of a malfunctioning intake. Branch. The local area received 10.3 inches of rain- WSO Williamsport issued 90 special statements fall over the 4-day period, June 21-24, with a 14- during the emergency, prompting extensive preven- inch rainfall center at Milton, Pa. Extensive small tive measures. Most damage in the vicinity of the stream flooding resulted. The record main-channel city was caused by small stream flooding, the main flow, which included the tremendous upstream dis- channel being contained by dikes. Main stream dam- charges, was contained by dikes with the aid of age was extensive below Williamsport, however. sandbagging and sand boil repair. A flood watch Five deaths were reported in Lycoming County. was issued on the afternoon of June 20, about 12 The river gage survived the flood, but radio com- hours before the beginning of heavy rain. The first munication was lost early in the rise. Fairly com- flood warning was issued at 4:00 p.m. on June 21. plete reports were received by telephone. The prin- A crest of 35.8 feet occurred at noon on June 24. 19 SUSQUEHANNA RIVER AT SUNBURY, PA. STORM TOTAL BASIN PRECIPITATION = 7 44 INCHES 0.5- 40 ~ "~ ~ ~ ~-- ~---~_~--.-~1 ,,~-OBSERVED CREST-358 FT -241200 'N ~ ~ ~ ~ ~ 1-."' ~ PREVIOUS RECORD-34 6 FT, 1936 -~~~ i-i, .,--$ r . C0- c-~ FLOOD STAGE-24 FT NO OBSERVATIONS AVAILABLE 231900 TO 251200 C~ 20- 10 &FLASH FLOOD WARNING 0 PREDICTED STAGE H---FORECAST TO STATED TIME I I ~I I [ I I 0 20 21 22 23 24 25 26 27 DATE (JUNE 1972) This was 12 feet above flood stage and 1 foot above Clingerstown, Pa. (about 30 miles north northeast the previous maximum in 1936 of Harrisburg), where the rain totaled 19 inches. A usable crest forecast was issued 36 hours be- Harrisburg was placed under a flood watch on the fore the crest arrived The principal protective afternoon of June 20. A warning was issued 24 works were the dikes, although all of the people be- hours later A sharp rise began at 1 a m. on June hind the dikes were evacuated Property damage in 22 At 2 p m, the 17-foot flood stage was reached. and around Sunbury was caused primarily by small, At 5 a m on June 23, the 1936 flood crest record of stream flooding Major main-stream damage oc- 29 2 feet was exceeded. The water was rising at the curred upstream and downstream from the dikes rate of 0 6 foot per hour at this time. The crest of The river gage was not destroyed, but telemetering 32 6 feet was reached at 1 30 a.m. on June 24. capability was lost owing to local power failure However, during the 16-hour period from 6 p.m. on during the evening before the crest Readings for June 23 to 10 a.m. on June 24, the level was within an undetermined period prior to the failure were one-half foot of the crest stage. A crest forecast of 1 5-feet low because of slippage of the telemark 32 feet was issued at 9:30 p.m. on June 22. float cable Loss of power to the RFC computer did On Monday, June 20, 15 key executive personnel not take place until after the principal crest fore- of the statewide civil defense organization began a cast had been issued week-long conference in State College, Pa. Based Susquehanna River at Harrisburg, Pa. on the flood watch issued by Harrisburg RFC the In the Harrisburg area, rain began on the morn- next day, this conference was terminated and the ing of June 21 and continued for 3 days, accumulat- personnel returned to their duty stations. ing to 15 inches. The main storm center was near Evacuation in the Harrisburg area was started on 20 SUSQUEHANNA RIVER AT HARRISBURG, PA. =0.X ITS gz:0g4-- PRECIPITATION AT CAPITAL ; ,,/CITY AIRPORT, HARRISBURG STORM TOTAL=15.8 INCHES MEAN PRECIPITATION FOR : BASIN ABOVE HARRISBURG i iSTORM TOTAL=8.1 INCHES Cp "i o~ ;S; L? A s--&-- OBSERVED CREST-32.6 FT.-241000 -32 tti i 000;< � � &0 v, v, vIAFLASH FLOOD WARNING /i r cr a~ *PREDICTED STAGE PRVU RiEO de d d l � �l_---FORECAST TO STATED TME-30 PREVIOUS RECORD / P T T 30 29.2 FT.-1936 / ~ v _ m \--28 7D~~0 0 - - ~ ~ I I ! ' .\DISCHARGE AT HARRISBURG - 300- FLOOD STAGE-17 FT.6 0 31 0_.6;lo, .t0. <>t / ad uDISCHARGE CONTRIBUTION OF '-,i-0/x LOCAL AREA BELOW SUNBURY O f'. 21% OF RUNOFF o-- 4- .. -- -I I I-. .- ..-2 : < X ~ p�20 21 22 23 24 25 26 27 DATE (JUNE 1972) the evening of June 21, because of local flash flood, stage. Subsequent river readings were made by RFC ing to 15 inches. The main storm center was near personnel using the wire-weight gage on the Walnut ing. Frequent statements issued by RFC, beginning Street bridge. When the bridge became inaccessible, with the 4 p.m. warning and continuing through the a gage was improvised. night, called for greatly worsening conditions and A 32-foot flood stage was forecast before power disastrous urban flooding from locally heavy rains. to the RFC computer failed, but considerable man- These warnings hastened and increased the scope of ual updating computations had to be made later. the evacuation operation. Property damage was Because the principal storm center was near severe in the city and surrounding area. One drown- Harrisburg, runoff from the portion of the basin ing occurred in the Harrisburg area. immediately above the city was much heavier than The telemetering capability of the Harrisburg that from the basin as a whole. The result was a river gage was lost early in the rise owing to radio very steep rise and an early peak from local inflow. failure. Later, the local observer had to be evacuated Near-maximum stage at Harrisburg was reached because of urban flooding from local rains, at a time many hours in advance of crests at upstream sta- when the Susquehanna was still 7 feet below flood tions. Although forecasts for points along the river 21 JUNIATA RIVER AT NEWPORT, PA. 2.5- 2 2*- STORM TOTAL BASIN PRECIPITATION=889 INCHES m o 1.5- z 0.5-_ 40 I- OBSERVED CREST-33 9 FT -232200 RECORD-34 2 FT -1936 30- - / ('" / NO OBSERVATION AVAILABLE _\ 222200 TO 251900 20 _- FLOOD STAGE-20 FT 20- A FLASH FLOOD WARNING 10-- 0 PREDICTED STAGE '---FORECAST TO STATED TIME 0 20 21 22 23 24 25 26 27 DATE (JUNE 1972) correctly predicted this somewhat unusual response, becoming heavy by night and remaining so all some recipients were reluctant to accept these fore- through the next day. The storm total for the basin casts, being accustomed to the more typical case of was 8 9 inches, of which 6 8 inches fell in a 24-hour a crest moving downstream from point to point. period A flash flood warning was issued at 3 p.m Paxton Creek flows through the center of Harris- on Wednesday, June 21, just as the rise started. burg. The numerous small-bridge openings, together The river continued to rise for the next 55 hours, with intense local iunoff, created a series of fixed- passing the 20-foot flood stage the following eve- orifice reservoirs. This effect aggravated the local ning and cresting at 10 p.m on Fiiday, June 23. flooding condition and produced flash-flood stages The maximum stage was 33.9 feet, just 0.3-foot less (min the city and near the confluence with the Sus- than the 1936 record. Precautionary measures evl- quehanna) fully as high as those which occurred dently consisted primarily of evacuation of resi- later from Susquehanna River backwater. dences. Approximately 700 people were removed, Juniata River at Newport, Pa. and a small army of volunteers moved large quan- The first flash-flood watch for this area was issued tities of furniture to high grounds. Property damage at 4.15 p m. on Tuesday, June 20, about 12 hours was heavy, but deaths were not reported in Newport before the beginning of rain. The rain began, at a or in Perry County. moderate rate, very early on Wednesday, June 21, Reporting service from the gage ceased at 10 p.m. 22 on Thursday, June 22, 24 hours before and 10 feet The Allegheny is noted as a fast-rising river when below the crest. Observations were not available heavy rains occur over the headwaters. The problem operationally until 2 days after the crest. The reason was compounded in this flood by general and con- for the initial loss of communication was the evac- tinuous very heavy rains. As a result, the main stem uation of the observer from his home, where the of the Allegheny rose like a flash flood, contributing remote river gage readout terminal was located, to short lead times in the warning issuances. Later, the river gage itself became inaccessible. Although official flood warnings provided limited Operational precipitation and river-stage data were lead time, and these warnings were generally issued lacking during the last half of the storm all along the during the night hours, the public was made aware Juniata River, because of the evacuation of observers of impending danger by extensive dissemination of and the failure of radio communication with RFC flood news through newspapers, radio, and TV. Ex- Harrisburg. cept for communities in the headwaters of the Alle- Upper Ohio (Allegheny-Monongahela) Basin gheny which suffered from flash flooding, there was In southwestern New York State, flash flood watches time for most people to take preparedness actions. were issued well in advance of flooding. Locl A major complaint from a marina operator in the wereissud wel i advnce f foodig. Lcal Pittsburgh area led to a Congressional hearing on knowledge of the flash flood characteristics of the Pittsburgh area led to a Congressional hearing on river in this area resulted in prompt community June 29, 1972. The President of the Waterways action and time ly evacuation of e ndangered areas. Association of Pittsburgh stated that the warnings action and timely evacuation of endangered areas. Because of the cooperation of county civil defense issued provided adequate time to take necessary and local safety officials, there were no iniuries and preparedness actions. The Mayor and Superintendent ~~~~~~no loss of life.of Police of Pittsburgh expressed concern regarding no loss of life. In Pittsburgh, flood warnings were generally avail- dissemination but otherwise were satisfied. There able with only a minimum of lead time, because were problems in the dissemination of information. of the suddenness of torrential rains and because When the mayor's office was called in the dissem- of the suddenness of torrential rains and because these rains occurred at night, when dissemination ination process, a telephone-answering device re- of warnings is most difficult. There was no loss of sponded. The Mayor stated that he expected the ~~~~~~~~~~life. ~Superintendent of Police to be fully informed at all ~~~~~~~~~life. ~times. The Superintendent said he solved the dis- River forecasts for the upper Ohio are prepared by semination problem by detailing one of his men RFC Cincinnati. Flash flood watches and warnings to the forecast office with a walkie-talkie. Once this are prepared and issued by WSFO Pittsburgh. had been done, he had no further problems. There Warnings are provided to the general public by was no loss of life. The local NWS Environmental the news media which obtain their information from Meteorological Support Unit (EMSU) was advised to the NOAA Weather Wire Service. The number of move from its location on the Monongahela River NWWS subscribers in Pittsburgh is limited. At the bank on Friday morning, well ahead of flood stage. major television station, the NWWS drop is in the The quantitative precipitation forecasts for the office of the staff meteorologist and not accessible river district, prepared as guidance material by the to others when he is off duty. National Meteorological Center, did not provide an Some commercial interests within Pittsburgh re- adequate picture of anticipated rainfall. Radar re- ceived warnings through a prearranged telephone ports on the NOAA Weather Wire did not reflect the alerting service operated by the Chamber of Comn- heavy rainfall rates during this event. Timely detec- merce. The Western Area Office of Pennsylvania tion of precipitation was not available from the Civil Defense received and relayed appropriate warn- normal substation reporting network. ings to country directors. Direct telephone calls were The local forecast staff did add additional rainfall made by the WSFO Pittsburgh staff to an extensive amounts to the model, based on some knowledge of list of community representatives. Warnings also rainfall in the area. These turned out to be too low were released to the Army Corps of Engineers, Pitts- in all cases where this technique was used. burgh District, which relayed these to its installations One trained hydrologist is on the staff at WSO along the Allegheny and Monongahela for local Pittsburgh, supported by two technicians to carry dissemination. out river district activities. This is not sufficient Flash flood watches for areas of western Penn- professional manpower to meet the hydrologic service sylvania were issued as early as 3:30 p.m. on June needs of this district. During the flood emergency, 20. Beginning on June 21, flash flood watches and the only hydrologist remained at his post 6 days warnings were issued frequently for numerous areas and 5 nights. It became necessary to assign a hy- over the upper Ohio. drologist from RFC Cincinnati to assist him. 23 ALLEGHENY RIVER AT OLEAN, N.Y. < 10 -- aSTORM TOTAL BASIN PRECIPITATION=9 08 INCHES c o ~ ~ o a � �- SPREVIOUS REC ORD 21 3 FT-1942 20-_ ...2O--A~o m a _n , &iLDTG1F/FLASH FLOOD WARNING 10- /O 1 PREDICTEOD STAGE-10 I--- FORECAST TO STATED TIME 20 21 22 23 24 25 26 27 DATE (JUNE 1972) Allegheny River at Olean, N.Y. Protective measures were timely and effective. The The rain began in this basin rather suddenly on the local civil defense office was warned at 8 45 a.m. evening of Tuesday, June 20. This rain was the Wednesday, June 21, and had 28 men on duty result of a frontal passage and deposited 3.5 inches within 1 hour. Precautions were not extensive until over the basin in the 12-hour period ending at 7 early Friday morning, when it appeared that the 25- a m Wednesday, June 31 A flash flood watch had foot flood wall might possibly be overtopped. About been issued at 3'50 p m on Tuesday, a few hours 6,000 people-one third of the city's population- before the rain began This was changed to a warn- were evacuated from 1,500 homes The National ing at 8 30 a m Wednesday, as a sharp river rise Guard moved both people and furniture The area began After the frontal passage, the rain stopped. evacuated was that which had been covered by the During the afternoon, moisture ahead of tropical 1942 flood which reached 21.3 feet Although the storm Agnes moved into the region, and another crest fell 8 inches short of overtopping the flood downpour began just as the river level passed the wall, there was considerable leakage and sewer back- 10-foot flood stage It lasted for the next 42 hours, up. Approximately 3,400 homes were affected. About adding 5 6 inches to the 3 5 inches already fallen. 90 homes located outside the flood wall were heavily The river continued to rise until 3 p m on Friday, damaged or destroyed The commercial area of the June 23, when it crested at 24 3 feet, 14 feet above city is not large, and had relatively light damage. flood and 3 feet above the previous record set in No deaths were reported. Throughout the event, 1942. radio station WHDL gave excellent cooperation and As the rain continued, several crest forecasts were provided a valuable service issued. Continued updating was necessary, but a sat- A network of four rainfall observers reports to the isfactory lead time was established. city of Olean, which relays the data to RDO Pitts- 24 OHIO RIVER AT PITTSBURGH, PA. NOTE PRECIPITATION AMOUNTS SHOWN (TOTAL 537 IN) ARE THOSE FOR THE 72 PER CENT OF "C' x u :THE BASIN WHICH CONTRIBUTED TO THE o 0_FLOW AT P/TTSBURGH THE RWMARNING 28 PER CENT DID NOT CONTRIBUTE DUE TO RESERVIR CONTROL TSE 232000 TOTAL WAS 585240700 24 25 26 27STRBUTED burgh. During the flood, telephone lines into Olean June 23, about 12 hours after the rise started. The directly to the RDO. The river gage is located at on Saturday, June 24. It was II feet above flood \ O ~gFriday , June 23. 00000 0 ' Ief j__OBSERVED CREST-35.8 FT.-240500 average of 5.5 inches of precipitation, spread fairly adequate during the most critical part of the opera-RNING ; .i'�:1��:"�~:��: . FLOOD AT RECORD-46 FT.-1936 / PREDICTED STAGE uniformly over a 5-day period. Twenty-eight percent tion. At times, calFLOOD STAGE-25 FT. TELEMETERING DEVICE OUT OF SERVICE 232000 TO 240700 200:00;tsn^05000 1 22 23 24 25 26 27 DATE (JUNE 1972) burgh. During the flood, telephone lines into Olean June 23, about 12 hours after the rise started. The became overloaded, and this portion was 5.85 inches. The repormaining weathed 35.-fooert crest was reached 22 hours nulater at 5 a.m. directly of t he RD. The river gage is located at on Saturday, June 24. It was 11 feet above flood the Olean sewage plant, which was the city's flood stage, but 10 feet less than the record of 1936. operation headquarters. Readings were made and A mash flood watch was issued at 4 p.m. on reported every 15 minutes throughout the rise. Thursday, June 22, and a warning at 2 a.m. on Friday, June 23. Ohio River at Pittshurgh, Pa. There were no failures of the telephone facilities The entire basin above Pittsburgh received an at the RDO, but the five lines proved totally in- average of 5.5 inches of precipitation, spread fairly adequate during the most critical part of the opera- uniformly over a 5-day period. Twenty-eight percent tion. At times, calls were coming in so fast that it of the basin is controlled by reservoirs which did not was impossible to place outgoing calls. This resulted release water during the rise. The average rainfall in complaints about the inability to contact the over this portion was 5.85 inches. The remaining weather office. People calling this number for river 72 percent of the basin received 5.37 inches and information were told, via a recording, that the produced all of the runoff that appeared at Pitts- office was closed. NWS received severe public criti- burgh. The greatest 6-hour amount which fell avo cism as a result. eraged 0.91 inches over the basin. The telephone lines to the river gage were sub- Although this was the largest flood ever to occur merged and failed at 10 p.m. on Friday, June 23, in June and although the concentration was down- and were out of service until late the following stream, the rise was only moderately steep. The morning. During the interim, manual readings were 25-foot flood stage was reached at 7 a.m. on Friday, relayed to the RDO by telephone. 25 HURRICANE FORECASTS City. The meeting resulted in a better common The National Hurricane Center (NHC) is re- understanding of the problems and needs of both sponsible for providing tropical cyclone forecasts and government and the public. attendant advice for the general public, marine and CENTRAL GUIDANCE PRODUCTS aviation interests, and the Department of Defense. The performance of the numerical models, which (Its responsibilities and operational procedures are represent the state-of-the-art, was generally adequate contained in the National Hurricane Operations in predicting the large-scale flow patterns. Accom- Plan.) panying the large-scale flow patterns was widespread Through the aid of satellite photos from the ATS- precipitation over the Appalachians and Atlantic 3, radar reports, and aerial reconnaissance performed coastal regions. The widespread nature of the pre- by the Department of Defense, the NHC accurately clpitation was forecast, and the unusual amounts located, defined, and tracked Hurricane Agnes. With forecast were helpful guidance to the field fore- the prediction computer models in use (primarily casters, in their issuance of the many forecasts of statistical), the NHC was able to confine the warn- flooding The excessive amounts which caused the ing zone along the Florida coast to about 100 miles. record floods, such as in the Susquehanna River The average warning zone for previous storms has Valley, fell in patterns of very small horizontal been about 250 miles. The average length of havoc extent compared to patterns now capable of being for each storm has been about 50 miles. In this predicted Figures C1 and C2 of appendix C, illu- case, there was complete destruction of buildings strate the fine detail in the excessively heavy rainfall. along the coast for a stretch of 20 to 30 miles. The landfall and early stages of Agnes were well Havoc along coastlines from hurricanes stems forecast both by the National Hurricane Center from two sources-winds and storm surges. Usually, (NHC) and the National Meteorological Center storm surges are the more devastating In Hurricane (NMC). Landfall is the direct responsibility of Agnes, the winds were below hurricane strength at NHC. Three NHC models provide numerical guid- the coast. Storm surges, forecast to be 7.8 feet at ance on landfall. a statistical estimate of trajectory; Apalachicola, were observed at 6 4 feet at a Coast a barotroplc steering model; and an analog estimate Guard station 3 miles east. of tralectory. All models are currently showing a The average accuracy of 24-hour predictions of similar degree of accuracy min forecasting the tra- hurricane landfall is a little over 100 miles. NOAA jectory over water as well as landfall. A typical plans to reduce this average to 75 miles over the 24-hour landfall forecast accuracy is about 100 next 5 years In the case of Agnes, landfall occurred miles. In the case of Agnes the excellent forecast of within 50 miles of the 24-hour prediction. landfall was accurate within 50 miles The accuracy Winds associated with Agnes were predicted to be of wind velocity and surface pressure estimates at about 85 m ph. with gusts up to 120 m ph. as the the time of landfall was not studied here as no storm penetrated the coast This forecast was based numerical model is involved on a value of 75 m p h reported by a reconnaissance Following a hurricane landfall, NMC is the sole flight at a location 25 miles off Cape San Blas. An source for guidance Throughout this storm's his- 85 m ph wind does not cause much public con- tory, however, an active dialog between the centers, cern, but a 120 m p.h wind does. The news media NMC and NHC, was continued. The NMC pro- picked out the latter figure, and used it in broad- duces the Central Computer Guidance for the casting warnings The highest gust reported was 55 Meteorological System. The models used by NMC m p h. at Apalachicola. Use of the high wind speeds are the Limited Area Fine Mesh Model (LFM) and by news media caused unwarranted concern by the the Primitive Equation Model (PEP) local populace and frightened away the tourist trade. There is no numerical guidance other than the To further compound this problem, Tyndall AFB LFM and PEP for hurricanes over land. Neither announced through the press early in the afternoon LFM nor PEP were designed specifically for hurri- of June 19 that the storm was over Albany, Ga., canes, but each handled Agnes well. and that military personnel were recalled to duty. The unusually large size of Agnes was a necessary The NHC warning, however, was in effect until condition for the numerical models to forecast it 6'00 p m. The NHC warnings served to keep tourists successfully The LFM was designed to forecast away another day. smaller scale phenomena than PEP, but it is too Apparent "overwarning" by NHC and consequent early to judge how well it will handle hurricanes economic losses by the local merchants led Con- generally, most of them being smaller than Agnes. gressman Sykes to call upon senior NWS officials The period of special concern in the flooding to meet with the concerned officials of Panama caused by Agnes extends from 1200Z on June 21 26 through 1200Z on June 23, 1972; from the time NMC forecasters successfully modified some the surface low pressure began to deepen over the characteristics of the numerical precipitation guid- Carolinas until the storm dissipated over Penn- ance by increasing the amounts and inserting detail sylvania. in the mountainous regions. They did not succeed, In general, the 500-mb numerical progs forecast however, in reducing placement errors in maxima well the large-scale evolution of the extra tropical to below the dimensions of mountain river and system initially over the Great Lakes and its rela- stream valleys, nor in forecasting the magnitude of tion to the smaller tropical system, Agnes. The 12-hourly accumulations. deepening of the extratropical system was well The relation between rain forecasting and signaled by both the LFM and PEP, including the flood forecasting. associated counterclockwise rotation of Agnes about The Quantitative Precipitation Forecast (QPF) the extratropical system June 21 to 23, and their is only one factor in flood forecasting. It is important final merger into one large extratropical system on to understand that in many cases the time lag be- June 23. Throughout the forecast period, the nu- tween onset of heavy flooding rain in a basin and merical models did a good job of predicting the the resultant flood crest exceeds the period of the position of the surface center of the tropical storm, weather forecasts discussed here. For example, WSO even to the extent of forecasting the storm's position Richmond issued its first flood warning for the over water at 1200Z on June 22 and to recurve in- James River more than 12 hours after the onset of land at 0000Z on June 23. The model consistently heavy rain but still 2 days before the flood crest. missed the deepening of the low-pressure center A flood watch was issued for Sunbury, Pa., on the throughout the period of intensification. Susquehanna River 12 hours before the onset of In terms of large-scale guidance the models per- heavy rain and 4 days before the flood crest. In formed well during this period. In a rapidly develop- such instances the rain forecast is more important ing baroclinic system of this type, it is typical for in terms of indications of continuance of heavy rain, the PEP model to move the system too slowly and but hard information such as from rain gages and underpredict the central pressures and attendant river gages plays the predominant role. In fact, it is gradients. Underprediction of the storm's intensity fair to say that generally such hard information plays may also have been caused by inadequate vertical the predominant role in today's flood forecasting resolution, particularly below the 500-mb level, system because of deficiencies in QPF. The im- Precipitation portance of improved QPF can hardly be over-em- Present skill in quantitative precipitation fore- phasized. Disaster prevention in the face of flood casting (as opposed to forecasting occurrences) lags stages requires actions which are more effective the well behind skill in forecasting flow patterns espe- longer the lead time of the watch or warning. Ac- cially in regard to numerical guidance. Quantitative curate QPF could extend the lead time. information on the LFM is just beginning to emerge. DATA COLLECTION This model clearly shows some skill in predicting In addition to surface and upper air observations, 0.25 inches or more of precipitation, especially dur- forecasts of hurricane movement and intensity are ing the winter and in the more northern areas. It based on observations by radar, reconnaissance air- is the general consensus that positive skill is a re- craft, and satellites. Specialized data required for flection of large-scale, well-defined synoptic features. predicting floods and flash floods are obtained by The excessively heavy rainfall which fell in the Sus- precipitation and river gages, radar, and satellites. quehanna River Valley was smaller in areas than Satellites the features predictable by the models. Information received from several environmental Although the state-of-the-art is not yet up to nu- satellites was used in tracking and forecasting Hurri- merically forecasting excessively heavy rainfall that cane Agnes from the storm's inception to its demise. falls in small-scale patterns, it does produce smooth These satellites are the geostationary Applications patterns that are generally regarded as useful by Technology Satellite, ATS-3, and the polar-orbiting practicing forecasters. During the latter days of the satellites ESSA 8 (APT) and ESSA 9 (AVCS). Nor- Agnes episode, 1200Z June 21 to 1200Z June 23, mally, one observation per day was available from the forecast patterns were generally forecast too far ESSA 8 (about 1600Z) and from ESSA 9 (2200Z). north and east, but the areas of observed 12-hour The ATS and APT pictures were available at NHC accumulations and their forecasts mostly overlapped. in real time. Still, the placement error was typically a couple of Nearly continuous viewing was provided by ATS-3 hundred miles or so, which is not unusual for to- from about 1200Z through 2100Z. Because of the day's models. storm threat, the viewing period was extended to 27 nearly 2200Z on 6 days (June 16, 17, 18, 20, 23, ings are based on the judgment of the observer or and 25) during the existence of Agnes and to nearly forecaster and confirmd by rain gage observations. 0000Z on 3 days (June 19, 21, and 22). On June Perhaps the most significant use of radar in flash 22, the satellite operated until nearly 0100Z. The flood forecasting during the passage of Agnes was normal operating mode calls for the satellite camera by WSFO Washington. Based on intense returns to scan one full earth disc m about 24 minutes The received by the WSR-57 at WSMO Patuxent, WSFO northern half can be scanned in half that time (12 Washington issued a flash flood warning for the minutes) when coverage of only Northern Hemis- Washington area and called for the evacuation of phere areas is required This mode was used on 7 Four-Mile Run. Flash flooding did occur a few hours days (June 18, 19, 20, 21, 22, 23, and 25), after a later, causing severe damage but no loss of life. series of about 10 full discs was received in support Radar systems min the areas affected by Agnes of Southern Hemisphere requirements. performed well throughout the period. Equipment Meteorologists at NHC and the National Environ- outages were minimal, extra staffing was called in mental Satellite Service (NESS) Analysis Branch when needed, and the radar data were utilized ex- conferred frequently during the storm, exchanging tensively in tracking the storm center and to some information derived from the satellite data. When extent monitoring the precipitation pattern. The the storm entered WSFO Washington's area of re- radar reports were especially useful to NHC as sponsibility, personal contact was made between Agnes approached the Florida coastline. Use of radar NMC and the WSFO Washington meteorologists information in predicting floods and flash floods A system for classifying tropical disturbances varied extensively throughout the system. based on satellite data is used when a disturbance is New techniques for presenting radar data have located over water. The maximum surface wind been developed recently When these become opera- speed can be estimated by means of this system. A tlonal, they can contribute to improvements in the coded teletypewriter message (called Satellite flood forecasting system both in accuracy and re- Weather Bulletin) containing information derived liability. The first system is a video integrator and from satellite pictures is transmitted once a day to processor (VIP), designed for use with the NWS field stations. Five such messages were sent concern- WSR-57 radars to display contours of equal echo ing Hurricane Agnes when it was over the Gulf of intensity on the radar precipitation map. This pro- Mexico-on June 14, 15, 16, 17, and 18. On June vides the operator and forecaster an immediate con- 18, the day before landfall, 65-knot winds were esti- toured map of the rainfall rates over an area of mated from the satellite pictures. When the storm about 125 miles radius. Of the 17 WSR-57s in the moved over land, satellite data continued to be useful storm area, only six stations had VIP available. in determining the extent of the most dense cloudi- Eventually, all stations will have VIP. ness and associated precipitation. Indications of on- To automate the flood forecast procedure, a going intensification and weakening, and some idea digitized radar data system (D/RADEX) is being of its motion, were inferred from satellite pictures. developed and tested. D/RADEX will have a capa- The satellite information used during Hurricane bilty of recording rainfall rate and transmitting the Agnes came from all the spacecraft sensors avail- data to a central computer where the cumulative able to NOAA at the time. These sensors use cam- rainfall by watershed area can be calculated. This era techniques for viewing only in the visible range. end product will be useful for flood forecasting. An If the sensors planned for the future satellites had experimental D/RADEX system had been installed been available, monitoring of the storm by satellite at the Buffalo WSR-57, but was not yet in opera- during both day and night would have provided a tional use during the period under consideration. A constant and continuous watch over the track and network of four of these systems is undergoing intensity of Hurricane Agnes. operational tests in the Midwest severe storm area. With D/RADEX, the operator will still be a key Radar link in flash flood warnings, since flash floods gen- Weather radar is used to estimate precipitation erally result from rapid developments in small areas rates by measuring the intensity of the reflected signal from precipitation particles. River and Rainfall Reporting Networks The radar system used by NWS for flood and Prediction of river stages depends directly upon weather forecasting purposes is the WSR-57, which observations of areal rainfall and of current and past was designed in the 1950's. Flash flood warnings river stages. Reporting networks of about 5,000 gages are Issued when the observer or forecaster observes supply this information daily and on a near-real- on the radar unusually heavy rain, or moderate time basis. These are operated and maintained by rainfall over an extended period of time. These warn- the various River District Offices. 28 About 95 percent of the river gages in these net- on the accuracy and reliability of the forecasts. These works are owned and operated by the U.S. Geo- forecasters had all worked long and hard under logical Survey or the U.S. Army Corps of Engineers, considerable pressure. Loss of reliable input data and are located by the owner agencies to serve their could only have the effect of further impairing their own missions rather than river forecasting. NWS effectiveness. Rainfall accumulations and river stages has access to these gages either by telemetering de- were increasing at such a rate that the usual cycling vices or through citizen observers. (6-hour) of data collection and forecasting was Rain gages are, for the most part, owned and inadequate. located by NWS. Some of the NWS gages are located The use of radar for precipitation determination and operated by the Office of Hydrology, some are is discussed in the radar section. However, coordi- a part of the meteorological synoptic and hourly nated effort between the radar program and the networks, while others are located and operated for rainfall network program is required to improve climatological purposes. These are supplemented by precipitation reporting. gages observed by cooperative citizen observers. The FLOOD AND FLASH FLOOD FORECASTS, location of gages depends on the availability of W ATCHES, AND WARNINGS WATCHES, AND WARNINGS observers somewhere near the most desirable site and the availability of telephone service. The forecasts, watches, and warnings issued by anithe aalaty fe texeptones, etra ie the field units of NOAA are based to a large extent With only a few exceptions, data are transmitted by commercial telephone. A few gages are served upon guidance products issued by NMC and sub- sequently observed conditions. The guidance prod- Osri ntok by radio. a ucts and data-collection systems have been discussed Observing networks, both river and rainfall, are invariably used by two or more agencies and the in previous sections. This section will deal with the bulletins-especially the flood forecasts and flash data relayed by a myriad of systems. In some in- stances, the citizen observer telephones data to two flood watches and warnings-issued by the field nors three agencies. In others, the information is units with emphasis on their timeliness, accuracy, or treeagenies In thes, te iformtio isand usefulness. A complete description of organiza- collected and r elayed by one agency to the others. tional responsibilities and definitions of the various Cooperation among the user agencies, from funding types of bulletins are contained in appendix A. A to data sharing, is necessary. The Agnes floods once again dramatized the weak- compilation of all bulletins issued is published in NOAA Technical Memorandum EDS NCC-1. nesses of the data collection network. In the case NOAA Technical Memorandum EDS NCC-1. of river gages, the cooperative observer usually The first action of the flood and flash flood warn- of river gages, the cooperative observer usually lives or works near the river. Not only does he have ing system was the issuance by the WSFO Washing- ton, D.C., Weather Service Forecast Office of a flash telephone lines vulnerable to flooding, but in serious ton, D.C., Weather Service Forecast Office of a flash floods he understandably is more concerned th flood watch for parts of West Virginia. This watch floodsaving his family and property than he is with was issued at 4:00 p.m. on June 19, while Agnes savig hi famly nd popery tan h is ithwas still over Panama City, Fla. Flood watches and reading the river gage. However, many cooperative was still over Panama City, Fla. Flood watches and warnings were extended the next day to cover parts observers continued observing and reporting at great of Virginia. On We dnes day, Jun e 21, as Hurricane persnal nconenince nd rsk hrouhoutthe of Virginia. On Wednesday, June 21, as Hurricane personal inconvenience and risk throughout the Agnes deepened and began its movement up the Agnes flood episode. Rainfall observers, in many e a s t coast, a series of flash flood watches and warn- instances, had difficulty in reporting due to telephone ings for New Jersey and eas tern, central, and north- failure, ~~~~~~~~~~ings for New Jersey and eastern, central, and north- failure. western Pennsylvania was disseminated by all NWS At one time, all reporting river stations on the offices in the area. At 6:00 p.m. that evening, a Schuylkill River in Pennsylvania were inoperative bulletin from WSFQ Washington stated that flood- for one reason or another. Seven of the nine river ing was expected to be near record levels on large gages on the James River were inundated. About 30 streams in the Carolinas and Virginia Wednesday percent of the rainfall reporting network in the night and farther northeast on Thursday, June 22. Harrisburg River District was not reporting owing From June 21 through June 25, flash flood and to communications failures, flood watches and warnings were issued in profu- River forecasts continued to be generated despite sion for specific localities throughout the five-State missing (and sometimes conflicting) reports. The area of Virginia, West Virginia, Maryland, Pennsyl- operations of the River Forecast Center were se- vania, and Delaware. Figures 2, 3, and 4 show the riously hampered. Precious time was devoted to areas where watches and warnings were in effect obtaining essential information, some of which was on June 21, 22, and 23. of doubtful quality. Forecasters were fully aware of Flash floods, by their nature, give very little time the deficiencies of the available data and their effect for warnings, and the warnings are primarily de- 29 Figure 2--Watches and Warnings In effect Figure 3.-Watches and Warnings In effect at 10 30 a m EDT, June 21, 1972 ; at 4 30 a m EDT, June 22, 1972 PORTLAND PORTLAND *.~~~ /~ ~~~~~F.F WATCH 'SMALL CRAFT FLOOD BUFFALO FF ALBAN BOSTON gA . ~ ~ WARNINGS WARNINGS-- WRNNGF BUFFALO < ,ji-TCH FL O OD F F. WARNING CLEVELAND * WATCH CLEVELAND C PITTSBURGH@ PHILADELPHIA \ PITTSBURGH PHILADELPHIA t FT WARSHINGTON WATCH WARNING ROUGH SEAS CHARLESTON WASHINGTON CHARLESTON HEAVY SURF I WARN I NGS F.F WARNING 0 RALEIGH ) FLOD FLOODS COLUMBIA cCO LUMBIA ' signed to save lives. Consequently, the lead times The primary method of disseminating NWS fore- for the flash flood warnings were short Some corn- casts and warnings to the public is through the munities on tributaries received no warnings or only mass news media, such as radio and television. those provided by alert local leaders. Radio and television stations acquire the mforma- The continuous heavy rains over entire river tion principally from the press wire services. The basins caused main stems to rise rapidly and, In NOAA Weather Wire Service and telephone calls effect, become somewhat like flash floods This un- to radio/TV stations also are used to keep the media usual condition contributed to reduced lead times informed about latest developments, NOAA Weather in the river flood warnings. Wire Service is a direct teletypewriter circuit from Nevertheless, NWS bulletins triggered public dis- weather offices to subscribing news media and other aster preparedness agencies into action throughout interest groups requiring the latest information. the five-State area The coordinated actions of these In addition, NWS uses automatic telephone an- agencies kept loss of life to a minimum For ex- swering devices, direct broadcasts on commercial ample, close cooperation between the NWS River radio stations, manually answered telephones, con- Forecast Center at Harrisburg and the Pennsylvania tinuous VHF-FM radio broadcasts, and indirect State Civil Defense office almost certainly saved channels such as State and community action agen- many lives The timely evacuation of 80,000 to cies, to convey warnings to the public Telephones, 100,000 people from Wilkes-Barre and the surround- both automatic and manual, and direct radio broad- ing area was an outstanding example of this team- casts, both commercial and Government-owned, are work the only real means of direct dissemination to the DISSEMINATION OF FORECASTS, WATCHES, public. AND WARNINGS During the east coast flood emergency, State A major problem encountered throughout the action agencies were served by direct telephone calls affected area was NWS's inability to communicate from the weather offices and by the NOAA Weather warnings directly to the public in a timely and Wire Action agencies relayed NWS information to effective manner. their community offices by internal teletypewriter 30 Figure 4.--Watches and Warnings in effect Figure .-Watches and Warnings in effect Press wire services gave "bulletin" status to most at 4:30 p.m. EDT, June 23, 1972. of the warnings. However, delays occurred between release of warnings by the NWS and their broadcast by the radio/TV stations. Time is required for re- typing at the press wire office, transmission over SMALL CRAFT the press circuits, receipt, and reading by the radio/ WARNINGS TV announcer. Previous episodes have shown that delays exceeding 1 hour can occur between a re- LAN BOSTON lease by NWS and eventual reading by radio/TV announcers. The public's radio/TV listening habits and broadcast schedules also kept many of them FLOARG NEW YORK unaware of the situation until they heard the warn- ings during the late evening or early morning news *TS HSLCRAFT broadcasts. Such delays are very critical in "short- WARNINGS fuse" events, such as flash floods. --*-L.~ ~Despite the built-in delays, radio and television CHARLESTON.- WASHINGTON ~stations did a good job of disseminating the fore- WARNIN~~~~G ~casts and warnings during the Agnes episode. The urgency of the situation required a substantial portion of the public to have direct access to flood information on an "on-demand" basis. This need can best be served by either telephone or continuous radio broadcasts. Within the primary flooded areas, only eight persons out of each 10,000 could be served by publicly listed, manually answered phones at NWS offices. One of the VHF-FM continuous weather broadcast sites is located at Washington, D.C. The tone-alert signal demuted special radio re- ceivers in the Washington area to forewarn of flash flooding. Activation of this tone provided the Dis- circuits and/or telephone calls. Isolated delays were ict of is eprg he i trict of Columbia's emergency headquarters with Asbtnoted inubr the relay wprocess. one of the initial warnings of the seriousness of the A substantial number of timely warnings were situation. Manual telephone calls placed by NWS tranmittd o theNOA Weaher ireServcesituation. Manual telephone calls placed by NWS transmitted on the NOAA Weather Wire Service, offices to special interest groups caused numerous but delays occurred before the warnings were broad- problems. Call-up lists are extensive, requiring sig- cast to the public. In the area affected by this dis- Cl- l s s t ete nse insi nificant times to complete. In some instances, they aster, NOAA Weather Wire is available on a state- appear to exceed the NWS staff's ability to respond wide basis only in Pennsylvania, where the percent- even under normal conditions. The use of these age of the media is very low and concentrated in a fthe Pmeldia aIn ver lote af etatesd lists, although important in warning dissemination, the Philadelphia area. In the other affected states, drained manpower resources to such an extent that the Weather Wire is limited to the cities having a e mpor r s u ort ute een t other important high-prority duties were in serious weather office. Overall, less than 10 percent of the eor an ipit dtese sr i jeopardy. Many recipients of these calls received news media in the affected area had continuous ac- the warnings more than an hour after release. cess to the latest forecasts and warnings. Even where many stations subscribe to the NOAA Weather Wire, PUBLIC RESPONSE the Service does not appear to be the most effective Public response varied widely, according to the means to disseminate warnings to the public. De- urgency reflected by the mass news media, past expe- lays occurred before the advisories were actually riences, wording of the releases, and the length of broadcast. In some cases, forecasts and bulletins lead time. When radio/TV announcers conveyed a were actually overlooked by the subscribing station. feeling of the seriousness of the situation (which was For example, when the power failed at Harrisburg, generally more intense as the event began), the the WSO Williamsport transmitted Harrisburg's ad- public became more responsive. Wording such as visories and warnings on the Weather Wire. How- that employed by WSO Richmond-"Prepare for ever, radio and television stations overlooked these severe flooding-Do You Remember the Camille data and were unaware that they originated at Flooding of August 1969?"-prompted quick public Harrisburg. awareness and response. Past experiences, generally 31 related to the respondee's age, contributed to differ- deal. The system reacted remarkably well, consider- ences in response Older persons, living in areas ming the magnitude of the emergency. which experience occasional minor flooding, were The staffing patterns along the path of Agnes were prone to ". stand pat and stick it out," even if it not abnormal compared with the prevailing overall meant moving to the upper floor of their homes. manpower situation within the NWS (see Appendix Younger people living in the same areas appeared to C). be more likely to move. The NWS Manpower Utilization Staff has estab- Public response was tempered somewhat by con- lished standards for, and periodically evaluates and fusion and misunderstanding. Few people know the revises as necessary, "required" manning levels for difference between watch and warning, and between all NWS stations. This program of defining objec- flash floods and river floods. The watch/warning tive and effective manning requirements is an on- misunderstanding gave some the feeling that NWS going activity which came about through recommen- was "crying wolf," when in fact watches had been dations of the 1969 Bohart Report Compared to issued to alert the populace to possible warnings. these standards, and additional ongoing programs, Confusion between flash floods and floods caused overall staffing levels at the NWS stations in the poor response to river flood warnings issued after path of Agnes were 7.9 percent under the recom- flash flood waters began to recede. mended and required levels. Lack of knowledge by action agencies in several The demands of an emergency situation such as communities, or failure to translate warnings into Agnes are not often proportionally distributed among community actions, as in Washington, D C, caused stations in the system. The greatest demand for ac- problems The inability of the local action agencies tion in an emergency may fall on the lesser station and news media to equate forecast stages with with a small staff, where the understaffing of only potential flood damage areas delayed public response one needed position may work tremendous hard- Lack of community focal points and community ships on the station personnel. In these instances, preparedness plans was evident throughout the dis- the work must be performed and the personnel must aster zone. endure whatever hardships are necessary to accom- FACILITIES AND STAFFING plish it. However, their endurance may be at the Facilities expense of peak personal efficiency from tired indi- Inadequate heat and air conditioning are provided viduals who must also temporarily eliminate other during nights, weekends, and holidays to NWS of- important station functions because of the priority fices located in Federal Buildings In addition, WSO demands of the emergency. While large stations may and RFC Harrisburg experienced an extended elec- frequently be better able than smaller stations to tric power failure during the height of the episode. cope with the understanding of one or two positions This affected the offices' operations and welfare of during an emergency, their personnel also suffer a the staff Sanitary, elevator, and lighting facilities very similar fatigue problem when performing a key were all inoperative. role in an exceptional emergency as great as Agnes. Manpower NWS is currently developing several programs The Agnes emergency demanded especial effort designed to improve manpower systems and utihza- from all concerned, and NWS personnel throughout tion within the service. These innovations will help the system met the challenge by responding with NWS handle more effectively such unusual emer- exceptional motivation and unselfish devotion to gencies as Agnes. These new programs will also re- duty throughout the long grueling hours of the or- quire additional staffing over the next few years. 32 APPENDIX A. The Flood and Flash Flood Warning System FLOOD WARNING SYSTEM Flood Forecast Products Organization In pursuing their responsibilities, RFCs and The National Weather Service's (NWS) River and RDOs issue the following products: Flood Forecast and Warning Program functions a. Flood Forecast Bulletin. A flood bulletin, pre- through a two-echelon system of forecast offices. It dicting specific stages at specific locations, is issued is designed primarily to provide river stage and crest whenever flooding is imminent or existing. Flood forecasts that follow the observable causative event Forecast Bulletins normally originate at a RFC, be- by more than 4 hours. cause the formulation of specific stage forecasts re- a. River Forecast Center (RFC). The first echelon quires complex computations involving rainfall-run- of the system is the River Forecast Center, staffed off relations and/or river-routing techniques. Flood by professional hydrologists, where river and water bulletins/statements are issued at periodic intervals supply forecasts are prepared. The RFC processes as long as flooding exists, to keep the public in- rainfall and river stage data in order to prepare river formed. forecasts and warnings for primary points along the b. River Forecasts. Operational river forecasts are river system. Its products vary from crest forecasts prepared for specific points within a river system and a few hours in advance for small drainage areas to are issued in terms of stage, volumetric flow, velocity forecasts made days in advance for downstream of flow, or combinations of these. The time range points on large rivers. RFCs also provide guidance contained in the forecast may vary from a few hours information on river and soil conditions to other to several weeks. NWS offices for use in preparing flash flood alerts, c. Headwater Statements. Advisory or planning watches, and warnings. Figure Al shows areas in data on headwater conditions, such as basin rainfall the Eastern Region for which RFCs have been as- amounts required to produce bankfull stages or signed responsibilities, cessation of flow are issued as "Headwater State- b. River District Office (RDO). The area served by ments." These statements prepared by the RFCs RFC is divided into several districts. In each district, contain guidance material for use by the RDOs, to a Weather Service facility is designated as the RDO, meet situations in which forecasts and warnings are the second echelon of the forecast organization. This not otherwise immediately available. office is directly responsible for the end product. FLASH FLOOD WARNING SYSTEM The district offices maintain networks of observing Introduction stations that report river stages, precipitation NWS Operations Manual Chapter E-13 provides amounts, and other parameters as required. These the national guidance and procedures for establish- reports are collected and relayed to the RFCs. Fore- ing and issuing: (a) flash flood alerts, (b) flash casts prepared at the centers are then transmitted flood watches, and (c) flash flood warnings. This to the river districts for dissemination. Issuance of document provided for regional modification of these river forecasts and flood warnings to the general procedures where necessary. The following is taken public, specialized users, and media such as news- from the Eastern Region Operations Manual Letter papers, radio, and TV stations is one of the RDO's 70-30. principal functions. When conditions warrant, RDOs Flash Flood Watch Bulletin may issue preliminary warnings before detailed a. Objective. The purpose of a Flash Flood Watch forecasts have been prepared by the River Forecast Bulletin is to alert the public and cooperating agen- Center. Figure A2 shows the areas of responsibility cies to the fact that current and developing meteoro- for RDOs in the Eastern Region. logical conditions are such that the area covered by 33 Figure Al.-River Forecast Center Figure A2.-River District Office Hydrologic Service Areas Hydrologic Service Areas ROC 4<"' RIC* a ~~~~~~~~~~~~~-. ALB/~~~~~1 (RFC) BUFS/ 0':zD,,i ,teF H'D FWA (RDO) i ' B.~., ' H CENTRAL I. / REGION C OlI m, ~~~~~~~~~~~~~~~~RFC) T HAR0 , R, CAK % P HARe '.. ~HAR r R - (RFC)\ ek ',.. !' ,, WASH � RFC ~ CINC ~D[ ADO cm SEiCE HI ~~~~~~~~~~~~~~~~~~~ .... ms ~ ICe '1 VEANDAAI GVAT[NN GENRA RIG DIS SER ICNAVE TYS RDO- / ". RD0 / ,.,/',. /~ ///,/0 RIVER DISTRICT OFFICE (RDO) *///AREA NOT SERVICED BY RDD RFC ,~,~ � CAE ~ AREA SERVED BY ROD OUTSIDE ATL EASTERN REGION (SOUTHERN�REGION) RFC-RIVER FORECAST CENTER (SOUTHERN REGION) RVICEDREASNOT(RFC RIVER FORECAST CENTER ALSO RIVER DISTRICT OFFICE the watch is threatened by flash flooding of a dam- Flood Watch Bulletin, the RDO will consult with aging or dangerous magnitude. the WSFO(s) having responsibilities in the area to b Rationale. The earliest possible public alert to be covered by the Watch and with the RWC. In the likelihood of the development of serious or dan- addition, if the RDO is supported by an RFC, that gerous conditions is of prime importance. It will Center will also be consulted. enable action agencies and individuals to plan for Any NWS facility may call the RDO's attention evasive or protective measures in the event that a to a situation which it feels may warrant the issu- Flash Flood Warning is issued or a flash flood oc- ance of a "Watch" However, it will remain the curs without further warning. function and responsibility of the RDO to issue the c. Procedures Conditions that require a Flash Flash Flood Watch Bulletin. Flood Watch Bulletin and responsibility for issuance RDOs are uniquely qualified for this responsibility of the Bulletin follow inasmuch as they are the only NWS elements having (1) A Flash Flood Watch Bulletin will be (a) a good knowledge of existing and forecasted issued when river conditions in their area, (b) up-to-date knowl- (a) a combination of meteorological and edge and understanding of antecedent conditions of antecedent conditions indicate a good probability soil and vegetation, (c) access to the very latest (greater than 30 percent) that flash floods will de- river and rainfall reports from substation networks, velop in a designated area; or and (d) the procedures and equipment for effective (b) the sudden break up of an ice jam dissemination of public river bulletins. threatens persons and property immediately down- A Flash Flood Watch Bulletin will be issued at stream any time that, in the opinion of the RDO, the con- A Flash Flood Watch may be issued whether or ditions described above exist To be most effective, not it has been preceded by a Flash Flood Alert these bulletins should be issued before 4 p.m. local (2) Initiation of a Flash Flood Watch RDOs time if hazardous conditions are likely to develop are solely responsible for the issuance of a Flash within the following 18 hours. Flood Watch Bulletin Prior to the release of a Flash d. Content of the Flash Flood Watch Bulletin 34 (see example). The bulletin shall contain the follow- ing information: EXAMPLE: FLASH FLOOD WATCH STATEMENT * The geographical area covered by the bulletin. This may be described in terms of well-known HSDC ER HYDRO RWC river basins or in terms of sections or portions of ABEEWR and NYC states. * The effective time of the bulletin. The time need I Attn 14 30Z : not be expressed in terms of hours but wording Trenton such as "this evening and tonight" is quite ac- AA tioice Trentn ceptable. j01:300 PM EDT TuesdayApril 14, 1970 * The extent of the hazardous condition expected, Flash flood watch for central and northern New Jersey i.e., localized or widespread. and eastern PA for this afternoon and evening. * The relative magnitude of the hazard, i.e., mod- DC and Atlantic City radars report TheWashingtonS DC and Atlantic City radars report erate or extreme. (Conditions indicating only light an area of heavy to very heavy rainfall extending from flooding do not warrant the issuance of a Flash Flood Watch.) ta0eastern MD across southern DEL into southern NJ. * The time of issuance and the originating RDO. This area is moving north northeastward. It is tex- * The statement that further clarifying advisories or pted to pass across NJ and eastern PA this after statements will be issued and the expected time noon and may cause localized flooding. Persons living of this issuance. in normal low lying areas or areas subject to flash e. Dissemination. The RDO will make every effort floods should remain alert and keep tuned to their to ensure that the public in the threatened area is radio or TV. An additional statement will be issued informed. This will be done through the mass news laterthisafteroon. media (radio, television, and wire services), action agencies (State and local police, Red Cross, Offices of Emergency Preparedness, Civil Defense, etc.), and by contact with responsible individuals in the area concerned (Flash Flood network personnel or substation observers). Cooperating agencies such as the U.S. Army Corps of Engineers, U.S. Geological Survey, and EXAMPLE: FLASH FLOOD WARNING State Water Resources agencies, shall also be in- formed. ALCKT MI A A PITC 161645Z All NWS facilities in the affected area and in :::: : adjacent or nearby areas will also be addresses of NA tial Weari Servie WBF the Watch Bulletin as well as the experimental Re- 10 AM T June 161970 gional Weather Center and the Hydrologic Services Ftlsh Fld Waring Division in NWS Headquarters. This will include Radar stations which are not in the "Watch" area Flash flooding is occurring at Greensburg PA, Ligonier Radar stations which are not in the "Watch" area but have coverage within the area. SPA, and Latrobe PA. Radar reports indicate that heavy Dissemination will be by telephone, NOAA thunderstorms and showers are still being observed in Weather Wire (or local loops), or TWX to recipients the Loyalhanna Creek drainage and are expected to external to NWS. Internal dissemination will be by continue throughout the morning. RAWARC, IWX, TWX, or telephone. Residbnts in Westmoreland County should remain f. Cancellation of the Watch Bulletin. The bulle- alert to the possibilities of additional flash floods and tin may be cancelled by the issuing office at any time further rises in the streams. that it becomes evident that the threat no longer exists. The WSFO(s) and the RWC will be consulted before cancellation. All addresses of the Watch Bulletin will be notified of its cancellation. Flash Flood Warning Bulletin a. Objective. The purpose of a Flash Flood Warn- ing Bulletin is to warn the public and action agen- cies that flash flooding is in progress or is definitely imminent. 35 b Rationale. NWS has the responsibility of warn- � the location and movement of the flood producing ing the public of existing or imminent dangerous storm, if known. conditions so that persons in the affected area may Follow up statements or advisories by the RDO(s) take immediate action to avoid loss of life or involved will clarify, to the extent possible, the na- property. ture, extent, and probable movement of the flood- c. Procedures Conditions that require a Flash producing storm. These subsequent statements or Flood Warning Bulletin and responsibility for issu- advisories should be the result of coordination with ance of the Bulletin follow: any of the several NWS faclhties concerned. (1) A Flash Flood Warning Bulletin will be e. Dissemination. Every effort will be made by the (1) A Flash Flood Warning Bulletin will be lssming office to warn persons in the area of im- issued immediately (a) If flash flooding is reported issuing office to warn persons in the area of im- mediate danger This must be done by any means or (b) if precipitation sufficient to cause flash flood- y available Police or Civil Defense personnel in the ing is reported. affected area should be notified either directly or (2) Initiatlon of a Flash Flood Warning Bul- through the State police or other action agencies. letin Any NWS facility has the responsibility and Immediately after the initial warning has been authorization to issue a Flash Flood Warning Bul- given, the issuing office, if not an RDO, will notify letin The necessity for immediate action is so the RDO having responsibility in the area covered great under these conditions that any delay cannot by the warning The RDO will then issue all sub- be justlfied However, responsible care should be sequent statements and advisories concerning this taken to ensure that the reports of flooding or havy event All questions and requests for information rainfall are not spurious will then be referred to the RDO. d Content of the Flash Flood Warning Bulletin The RDO, upon notification that a Warning has (see example) Owing to the urgency and seriousness been issued in its area will notify affected WSFO(s), of flash floods, the initial warning bulletin need RDO(s), RFC(s), and the experimental Regional only contain the following' Weather Center of this action. In addition, the RDO � the location of the observed and reported flood- will to the extent possible verify that the warning ing or heavy rainfall; has reached the affected area and any additional - the streams and rivers being affected, if known; threatened areas. Further warnings, as warranted, * the magnitude of the flooding, if known, and will be issued by the RDO. 36 APPENDIX B. Winds During Hurricane Agnes The accompanying charts (figs. BRI through Ri II) winds were no longer a significant feature of the depict the position of Agnes and the areal extent of public forecasts, gale and/or small craft warnings wind warnings given in advisories for Agnes at 12- were continued in coastal areas, hour intervals. (Winds are 1-minute average winds as On June 21, Agnes increased in intensity, and reported in hourly observations. Gusts are peak the winds were once again significant, especially in winds observed within 10 minutes prior to the ob- the coastal marine areas where gale warnings were servations. Higher winds occurring between obser- in effect. With this intensification of Agnes, the vations are quite likely.) bulletins issued to the public and news media on Gale warnings and a hurricane watch were issued June 22 not only carried gale warnings, but also from Dry Tortugas to Key West, Fla., at 6:00 p.m. indicated winds of 45 mph with gusts to 60 mph to EDT June 17, when Agnes was still near western the immediate southwest of the storm center. The Cuba. At noon on June 18, a hurricane watch was Chesapeake Bay Bridge Tunnel reported a maximum issued for the upper Florida coast from Cedar 'Key wind of NW 57 mph with gusts to 69 mph. Wallops to Pensacola. At 6:00 p.m. on June 18, hurricane Island reported a maximum wind of NW 45 mph warnings were issued from St. Marks to Panama City with gusts to 64 mph. (Fla.) Beaches. Highest winds of 85 mph were re-Asgnsctiudomveorhate ported by NOAA reconnaissance aircraft on June maxiu winds cotended to doecrae Stormhwarn-th 18, while Agnes was about 275 miles south ofinswrisudfrtelerGatLka- Panama City. As Agnes approached the coast on togminimlstweremsse forcte widswere expectLaed.a June 19, a reconnaissance flight reported winds of The bormlyhgh liialkstormeforcemwineds with exethed 75 mph at a location 25 miles off Cape San Blas, wind tboral hihcause lexesvelshr frontdmbineduwit the Agnes crossed the coast near Cape San Blas, a short winst aus xesve battefryn dacaedetion distance southeast of Panama City, during the late wv atn cin afternoon of June 19. The highest wind gust re- After landfall on the coast of Florida, wind was ported by a land station in this area was 55 mph not the dominant feature of this storm. However, an at Apalachicola, Fla. examination of the advisories and bulletins issued Once over land, winds diminished rapidly: e.g., on this storm indicate that the public and coastal June 19, 6 p.m. advisory, "winds estimated 55 mph marine interests were well advised on wind condi- in squalls;" June 20, 6 a.m. advisory, "highest tions. Further, a comparison of forecast winds and winds estimated 35 mph;" and at noon on the 20th, actual maximum winds appears to be in reasonably "highest winds estimated at 25 mph." While the good agreement. 37 U;~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ : L :: 46 181817 ESTS / SA 2~1E NNE35 644 19030 ES I1032 S .~. *SE3 655 181816 EST ....... .6 ............ SE27 G43 191109 EST@ CIRT ID P1 ... ........... E~18132 G46 . .. 75 MPH AT 1500 FT. OFF 18132 EST46~ APPALACHICOLA 191233 EST / SE25 G53 180955 EST~ *E2 63 . ~~~181523 EST SSE42 G56 180700 ESTI SSE32 G50 181200 EST# dmbb' /~SE43 G52 180144 E T1 RN:iiWj 25-38 7[T3 7/ U 10011M.P.H. 111012-38 M.P.H. U dUU~~~~~~ ~~~/i$///~~~~~~39-54 M.PH. M.P.H . %.075 M.P.H. *7~ANS 53* 5 M.P.H N.OR HIGHERM*'s " ... OR HIGHER ---~~~~~~~~~~i-.-i;-� -: ""~ Figure SI.-Wind warnings includes in 6-a~m. Figure B2.-Wind warnings included in 6-p.m. advisory number 8, June 18, 1972. advisory number 10, June 18, 1972. 38 g~~~~~~~33 HIM~3 21 33 3~~~'3~~'3 ~~~U3 E20 200300 ESTIO *E25 192355 ET: NE22 I9XXXXOES NE 38 19XXXXO *E410X~~.~ * E24M 63 9204 EST E33 19xx 0E79X NE25 19XXXXO *NE25 640 10930 EST11E9 G56 190637 EST *NE25 643 192200 EST G49190740 ESTM ~ ~ *ESEO G4O 191042 EST . ~~~//////~~~~EE... ACFI RPTD O,AGNES , WINDS tip E1 TO 75 M.P-H AT 1500 FT. OFF P APPALACHICOLA 191233 25-38 M.PH. ........MPH. ....~.. 53 .. 39E 54 M .PH. OR5IGHR3 8 M.P.H. XON,~~~~~~~~~~~~~~~~~N Figure B3.-Wind warnings included in 6-a.m. Figure B4.-Wind warnings included in 6-p.m. advisory number 12, June 19, 1972. advisory number 14, June 19, 1972. 39 SE2B 201038 EST@~~........ ~~~~~~~~~~~... ... 2538~ ~~~~S2 MPH.3 2E-8SPH Figure B5.-Wind warnings included in 6-am. Figure B6.-Wind warnings included in 6-p~~~~~~. . .. . . . m. adioynmE206 Jn 20030 EST2 bultnJn00 92 E25 192355 EST~~~~~~4 W43 G E ST G*W45 64 220445 EST NW7 059 220130 ESTO BRIDGE TUNNEL NW42 5 212ET AGNE WNW37 062 212158 EST~N 62 211346 EST.'N 25-38 MP.H.238MPH "K~~~ OFFSHORE WINDS . ~~~~~~~~~~39-54 M.PH. N.X 39-54 M.PH.MM Figure B7.-Wind warnings included in 6-a.m. Figure BS.-Wind warnings included in 6-p.m. bulletin June 21, 1972. bulletin June 21, 1972. 41 E26 G39 22XXXX . E25 22XXXX* "-'...' -....W. :....:.: AB\:.:. SSE44 22XX\ S26 G40 222132 EST * S26 G38 221928 EST , ENE21 22XXXX E30 22XXX � S29 646 221956 EST . AG SSW37 221900 EST?: WNW35 G51 N WILMINGTON N21 23XXXX@ SW29 G38 221854 EST *W34 22XXXX NW31 G49 221225 ESTNW.49 221200 EST W30 G34 2210 ESTO HARRISBURG NW32 G50 210708 ESTNW40 G46 221200 EST NW27 :42 22XXXX N47 221200 EST S IAGNES NW40 221200 EST _ ::0r. NW45 655 220900 EST SW20 G50 220900 EST : N24 G31 221555 EST *: .....:.:.::.:.::..::: ..-. iiii iiiii25-38 MP. :. 25-38 M.P.H. 39-54 M.P.H. 03 39-54 M.P.H .. . Figure B9.-Wind warnings included in 6-a.m. Figure B10.-Wind warnings included in bulletin June 22, 1972. 6-p.m. bulletin June 22, 1972. 42 *SE29 231029 *S~__ 0 G63 23XXXX N4?.060!2725 -38 M.P.H. 4 2 X ABOVE 54 M.P.H. Figure B11.-Wind warnings included in 6-a.m. bulletin June 23, 1972. 43 NE2 ::::::: ~~�1.�;)�:,� �i,: ~~~~~~~~~~~~~~~~~~~~~~~~~::::: ' i- 1~- �i iiii 43i 2 2 22 (6 I ~6 ~~~/a 2 4 NJ 6 ~~~a 1J ~~~~~~~~~~~~~~4 _ F'igure Cl,-Total precipitation in inches dur- Figure C2.--Total precipitation in inches dur- ing Hurricane Agnes, June 18-25, 1972, ing Hurricane Agnes, June 18-25, 1972, for southeastern United States. for northeastern United States. , .....'""'"' --C'17.4 AHTTANGO CR. PA. (HARRISBUR,~ 38 Mi. NNE) / " ~~~~3.76 WASHINGTON, D.C. (DULLES) 13.77 WELLSVILLE, N.Y. .. 2(... .... ) /~~~~~~~~~~~~~~~~81 TALAHASE, FLAs.oo /~~~~~~~~~~~~~~~~55 GREENSBOR, N.C../i... 3.34 ATLANTA, GA. 18SUN. 19 MO. 20 TUES. �D 21V 2 THIN.: 23 FRI. 2 SAt. 25si Figure C3.-tCumul ative r a infall curves for selected locations during Hurricane Agnes, June 18-25, 1972. 44 APPENDIX C. Precipitation During Hurricane Agnes Storm rainfall for the period June 18-25, 1972, values were near the eastern Appalachian slopes. Of is shown by isohyetal maps (figs. C1 and C2). Rains particular importance were heavy rains of up to 6 of generally 4 inches or more extended from Florida inches near the central border line between New to New England and from the east slopes of the York and Pennsylvania. The curve for Wellsville, Appalachians to the coast. This isohyetal map is N.Y. (Fig. C3) shows this earlier rain burst. based on unchecked regular reporting precipitation For the next 24-hour period, ending the morn- data plus all supplementary measurements gathered ing of June 22, heaviest amounts were measured by survey teams and RDOs. The survey teams cov- along a north-south swath near central Pennsylvania. ered portions of New York and Pennsylvania. These Figure C3 shows cumulative curves for Washing- surveys were joint efforts of NWS and the Corps ton, D.C., and Harrisburg, Pa. (38 mi NNE). Wash- of Engineers. ington was deluged with over 11 inches in less than Figure C3 gives mass rainfall curves (plots of 18 hours. Harrisburg (38 miles NNE) had over accumulated rainfall with time) for selected recorder 13 inches in the 24-hour period ending near 6 p.m. rain gage stations. on June 22. For the 24-hour period ending the morn- For the 24-hour period ending the morning of ing of June 23, rainfalls occurred generally from June 19, while the storm was moving due north in Maryland northward. During this period, a second the Gulf of Mexico near the longitude of Apalachi- heavy burst occurred in central New York, as shown cola, rains averaging about 6 inches occurred over by the mass curve for Wellsville. most of Florida and south Georgia. A peak of 14.5 For the next 24-hour period rains covered ap- inches was measured at Titusville on the mid-At- proximately the same region. proximately the same region. The last day of the lantic coast of Florida and 12.9 inches at High storm period (ending on June 25) few rainfalls Springs near the border of Florida and Georgia. were greater than half an inch and were generally The curve for Tallahassee, Fla., is representative of from Pennsylvania northward the first day's rainfall. For the next day, June 19, rains spread over eastern Alabama, Georgia, South Of particular interest is the region covered by the Carolina, and into southern Virginia. Rainfall curves large eight-inch isohyet (36,000 sq. miles) centered are given for Atlanta, Ga., and Greensboro, N.C. in Pennsylvania, but extending into New York, (fig. C3). By the morning of the 20th, the storm Maryland, and Virginia. Average rainfall over this center was located near central Georgia, having area for the storm period was close to 11 inches. crossed the Florida coast near Panama City at about About 9 inches fell in 48 hours, and 6 inches in 24 5 p.m. on June 19. hours. For the 24-hour period ending on the morning of Rainfall within the 14-inch isohyet in Pennsyl- June 21, rains were widespread over much of the vania, covering about 3,000 sq. miles, averaged over eastern seaboard from Georgia to New York. Largest 15 inches; almost 11 inches of this fell in 24 hours. � U. S. GOVERNMENT PRINTING OFFICE :1973--511-324/213 45