GreenWorks Pennsylvania Department of Environmental Protection Pennsylvania Home Site
Home Page Subjects Search PlanetPA AskDEP Participate!

THE FLOOD OF JANUARY 1996 - A SPECIAL HYDROLOGIC REPORT


The flooding that took place throughout the Susquehanna basin on January 19th and 20th, 1996 was an unprecedented meteorological and hydrological event. Much of the stage for the flooding was actually set in the preceding months by above normal snowfalls and below normal temperatures, a regime that had persisted basinwide since the beginning of November 1995. Table 1 shows the snowfall totals in inches for the season at four "first order" National Weather Service stations in the basin. Table 2 shows heating degree day data and days with the maximum daily temperature below freezing for the same time period. These tables indicate that by mid-January, the winter of 1995-96 was well on its way to becoming a "record-setter". The most memorable event prior to the flood was the "Blizzard of 1996" which blanketed the southern two-thirds of the basin with a significant snowfall that ranged from 18 to 24 inches. Adding to this total was another storm several days later that added almost another foot of snow basinwide on top of "The Blizzard". Mid-January data indicated there was a snowpack throughout the basin that contained a water equivalent of from 3.5 to over 5 inches. This exceptionally deep snowpack, combined with the fact that there were very few opportunities for the snow to melt throughout the season because of the below normal temperatures, was certainly one of the factors that contributed to the Flood of January 1996.

Table 1

November
December
January
Total
ActualNormal ActualNormalActual Normalas of Jan. 19
Harrisburg, PA8.12.1 17.47.338.9 10.864.4
Williamsport, PA13.8 2.925.08.1 32.721.869.1
Wilkes Barre, PA18.6 3.315.98.6 37.512.069.9
Binghamton, NY29.27.3 32.717.728.3 20.086.0

Table 2

Number of Heating Degree

Days Above Normal
Number of Days Temperature Did Not Get Above Freezing
Harrisburg, PA
216
25
Williamsport, PA
331
37
Wilkes Barre, PA
144
38
Binghamton, PA
196
53

Another major contributor to the flooding situation was ice. The air temperatures through most of December 1995 and January 1996 were below normal, with very few periods ever getting above freezing. This caused the formation of ice on the river, up the mainstem Susquehanna as well as on some major tributaries throughout the basin. This ice cover would play a significant role in the flooding problems later in the month, especially aggravating the situation in the southern portions of the basin.

The region then experienced an unprecedented meteorological event on January 18th and 19th. Air and dew point temperatures rose from below freezing into the 30s and then into the 50s in a matter of 12 to 24 hours. Temperatures stayed above freezing for about 48 hours. Strong southerly winds gusting as high as 50 mph accompanied the rapid rise in temperatures. Sustained wind speeds varied, but generally 20 to 30 mile per hour averages were common through this period. These conditions caused a very deep snowpack "to ripen" very quickly. This ripening is the process whereby a deep fluffy absorbent blanket of snow becomes saturated as the high winds and temperatures cause the snow temperatures to rise to the 32 degree level and begin melting. This ripening precluded the snowpack from absorbing and freezing any of the rainfall that would follow and actually contributed an average of 1.5 inches of runoff from snow melt. This warm air and high winds was associated with a frontal system moving through the basin. This weather system focused heavy rainfall across the region during this event as well. Rainfall averaged 2.5 inches from this event, with most of the rain falling in just three hours. Basin average rainfall amounts were in excess of two inches throughout much the region, with the heaviest hit areas approaching three inch averages (Attachment 1). These meteorological conditions produced rapid excessive runoff throughout the region. The precipitation and associated runoff on January 19th has been described as a Susquehanna River basin wide flash flood event because of the intensity of the precipitation and the speed of the runoff.

With between three and four inches of runoff occurring almost simultaneously throughout the basin, stream levels rose quickly. This rapid rise caused any ice cover that was present to break apart. As the ice cover broke and was washed downstream, many ice jams occurred at bridges and natural constrictions on both the tributaries and mainstem. As more ice and water rose and backed up behind these jams, they broke, often with catastrophic results to those areas immediately downstream. Attachments 2, 3, 4, and 5 show the river stage hydrographs for the USGS mainstem gaging stations at Sunbury, Harrisburg, Marietta, PA and Conowingo, MD respectively. An examination of the Sunbury graph shows the "normal" rapid rise of the river apparently unaffected by ice jam formation. This can be contrasted with the Harrisburg graph which shows a definite effect of one or more ice jam formations on the rising limb of the hydrograph. These ice jams, when broken, send a surge of much higher flow downstream. This is what happened below Marietta at 1 p.m. on Saturday, January 20th and culminated in the flooding event later that evening in Port Deposit, MD. In addition to the four mainstem hydrographs described above, Attachment 6 shows discharge hydrographs for six stream gages located throughout the basin. The one characteristic common to all of the hydrographs is the very rapid rise from near normal conditions to flooding conditions in a matter of hours. The magnitude of this event is further emphasized by the tabulation in Attachment 7. This tabulation shows, for various gaging station locations throughout New York and Pennsylvania, the maximum average daily flow value recorded as a result of the flood. This average daily value is then "ranked" according to all the daily flows values ever recorded in January at the respective gaging stations. Finally, the median daily flow value for the third week in January is shown along with a ratio of the January 96 maximum to the median. In the Susquehanna basin, these ratios range from 12.7 to 36.3, with an average of 22.1.

The operators of both the Safe Harbor and Conowingo hydropower facilities were faced with a very unusual event on the afternoon of January 20. A large ice jam that had formed approximately five miles upstream of the Safe Harbor Dam in the vicinity of Turkey Hill broke at approximately 1 p.m. on January 20. The release of water from this ice jam caused the level of the river to rise very quickly behind the Safe Harbor dam. The operator of the dam began releasing water through the Safe Harbor flood gates, simultaneously notifying downstream utilities and government entities according to a predetermined action plan. The dam operator has estimated that the river flow below Safe Harbor was increased from 224,000 cubic feet of water per second (cfs) to 826,000 cfs in a two hour period. It must be noted that these flow values, particularly the higher estimates, have been made as a result of information gathered after the event. The flow values that were estimated during the event, based on the best information available at the time, were lower. This increase of flow into Lake Aldred and subsequently into the Conowingo pool happened very quickly, necessitating Conowingo operators to quickly institute their emergency operations plan. In order to control a constantly rising pool, Conowingo operators had to open a total of 42 flood gates before the situation was stabilized on the evening of January 20 (reference Attachment 6).

As bad as the Flood of 1996 was, it could have been much worse. The U.S. Army Corps of Engineers estimates that its flood control reservoirs stored a total of 167 billion gallons of water that would have otherwise added to the problem. This flow reduction, coupled with a wide array of local flood protection projects such as channels and levees, is estimated to have prevented a total of 1.37 billion dollars of additional damage. A tabulation of estimated stage reductions along with a more complete breakdown of damages prevented is included as Attachment 8.


BALTIMORE DISTRICT CORPS OF ENGINEERS

JANUARY 19-22, 1996 FLOOD EVENT

(PRELIMINARY DATA)

FLOOD DAMAGES PREVENTED



SUBBASIN
DAMAGES PREVENTED
$ millions
Chemung River237.9
Upper Susquehanna River112.3
West Branch Susquehanna River333.2
Juniata River70.3
Susquehanna River - Main Stem612.2
TOTAL1,365.9

FLOOD STAGE REDUCTIONS

NATURAL
OBSERVED
STAGE
FLOOD
SUBBASIN
LOCATION
STAGE
STAGE
REDUCTION
STAGE
Chemung RiverCorning, NY 33.1825.937.25 29
Elmira, NY18.65 13.255.410
Lindley, NY23.82 13.1810.6414
Juniata RiverLewistown, PA 39.3331.537.8 23
Newport, PA30.36 24.695.6722
Main StemDanville, PA 28.5425.972.57 20
Susquehanna R.Harrisburg, PA 25.5124.970.54 17
Sunbury, PA32.06 30.321.7424
Wilkes-Barre, PA36.31 34.222.0922
UpperChenango, NY14.73 12.492.2410
Susquehanna R.Owego, NY 44.1340.243.89 -
Vestal, NY30.23 27.82.4318
West BranchLewisburg, PA 28.6925.922.77 18
Susquehanna R.Lock Haven, PA 26.8423.833.01 21
Williamsport, PA31.44 26.714.7320