Hurricane Connie became the first hurricane of the 1955 Atlantic Hurricane Season on 5 August over the open waters of the North Atlantic. By 6 August, Connie had rapidly intensified, reaching peak intensity with maximum sustained wind speeds of 145mph and a minimum central pressure of 936mb. Connie retained category four strength for several days as it moved just to the north of the Leeward Islands and Puerto Rico.
As Hurricane Connie approached the Mid-Atlantic coastline, the system began to take a slow erratic turn to the north. In the process, Connie weakened substantially over cooler waters and eventually made landfall over the North Carolina Outer Banks on 12 August as a category one hurricane with maximum sustained wind speeds of 80mph.
Connie continued northward after landfall, moving through the Chesapeake Bay before turning to the northwest and passing over Pennsylvania and southwest New York. Connie eventually lost all tropical characteristics on 16 August, while over southern Canada.
Although Connie brought hurricane force winds and up to 8 feet of storm surge to the Carolinas, heavy persistent rainfall was the most significant aspect of this storm (Figure 1). A swath of 3.00 inches (7.62 cm) or more of rain covered much of the Mid-Atlantic and southern two-thirds of New England, with a maximum storm total accumulation of 13.24 inches (336.3 mm) reported in Fort Schuyler, NY. This rainfall resulted in localized flooding, and accounted for much of the $40 million in damage that was caused by Connie.
This paper will examine the large scale patterns associated with the heavy rainfall during Hurricane Connie. The focus will be on examining climatic anomalies (Hart and Grumm 2001; Grumm and Hart 2001) to determine if this storm exhibits synoptic behavior similar to that of other Mid-Atlantic heavy rainfall events (Table 1). Additionally, this paper will show that the significance of flooding in the Mid-Atlantic depends on much more than just precipitation.
The 850 hPa heights, 1000 hPa precipitable water, and other standard level fields were derived from the NCEP/NCAR data set. The means and standard deviations used to compute the standardized anomalies were from the NCEP/NCAR data as described by Hart and Grumm (2001). Anomalies were displayed in standard deviations from normal, as standardized anomalies. All data were displayed using GrADS (Doty and Kinter 1995).
The standardized anomalies are computed as:
SD = (F – M)/σ (1)
Where F is the value from the reanalysis data at each grid point, M is the mean for the specified date and time at each grid point and σ is the value of 1 standard deviation at each grid point.
The precipitation images and the ranking of the top 5 November rainfall events in Table 1 were based on the UPD data from 1 January 1948-1 October 2010. The maximum rainfall over the 24 hours ending at 1200 UTC daily was used to find these data. These data are heavily biased toward COOP reports, but they represent the longest running continuous record from which these records can be easily estimated. The events were classified based on the orientation of the frontal system. General north-south frontal systems are associated with the Maddox-Synoptic type rain events (Maddox et al. 1978). Tropical systems which interacted with north-south frontal zones were classified as hybrid events. Those which did not interact with fronts were classified as tropical events. Frontal events were classified based on a more east-west oriented frontal boundary and significant easterly wind anomalies.
The National Hurricane Center’s best-fit track database (HURDAT) was utilized to examine Connie’s track and intensity (Jarvinen et at 1984).
For brevity, times will be displayed in day and hour format such at 12/1200 UTC signifies 12 August 1955 at 1200 UTC.
i. Large-scale pattern
A dominant center of high pressure built in over the Great Lakes from the west on 12 August. This high slowed Connie’s forward speed as it meandered northward towards the Carolina coastline. Connie’s slow movement over an area of cooler sea surface temperatures (SSTs) in the north Atlantic, in combination with the dry air to the north of the system lead to the weakening of Connie from an intense Category 4 storm to a category 1 storm at landfall.
With this high pressure system in place, Connie did not have a chance to directly interact with any other low pressure systems. Therefore, unlike many tropical systems that impact the northeast, Connie was not quickly swept back out to sea by an approaching Maddox synoptic or frontal type system (Maddox 1979). In fact, Connie slowly dug northward across the Mid-Atlantic, before losing all tropical characteristics over the Canada and the Great Lakes on 15 August.
Figures 2 through 4 show the evolution of the general pattern as Connie made landfall and dug into the Mid-Atlantic Coast. Analyses of these figures explain how Connie was able to produce a significant amount of rainfall over the region. Figure 2 shows a maximum in precipitable water of three standard deviations above average over the Mid-Atlantic coast at 13/1200. By 14/1200, the center anomalously high precipitable water content had shifted into Canada. These high precipitable water amounts were coupled with large anomalies (of 5 standard deviations or greater) of 850mb pressure, 1000mb pressure, and easterly wind components. These factors proved to be ingredients for a historic rain event, as tropical moisture poured in over the Mid Atlantic for an extended period of time.
ii. Rainfall and flooding impacts
The estimated total rainfall is shown in Figure 5. A closed contour of 128 mm stretched northward from where Connie made landfall in North Carolina up into southern Vermont and New Hampshire. Within this wide area of high QPE, there were locally higher amounts of greater than 196mm.
Between just 11/1200 and 12/1200, there was a rainfall maximum of 193.849 mm in the Mid-Atlantic. No rain event in our record, which begins in 1948, has produced more rainfall in a 24 hour period starting at 12Z (table 1). Another location measured 149.511 mm of rain the next day. This shows that Connie continued to produce heavy rainfall over land even as it began to lose its tropical characteristics. In fact, the maximum storm total rainfall fell in Mount Schuyler, NY.
With such historic rainfall typically comes flooding. The Mid-Atlantic River Forecast Center reported that only 16 of its forecast points experienced flooding, with just one location, Neshaminy Creek in Langhorne PA, reaching major flood stage1. Furthermore, every single river gauge crested below flood stage in Virginia, where there was a local precipitation maximum.
Hurricane Connie was a historical rain event in the Mid-Atlantic and New England states. It is fairly typical for these historic Mid-Atlantic rainfall events to be associated with a tropical cyclone (TC), like Connie. Of the top fifty 24-hour rainfall events in our database, 30 of them were influenced by a tropical cyclone or their remnants.
Because of the geography of North America, TCs usually need to interact with a baroclinic feature to get enough curvature in their track so that they can move far enough north to affect the Mid-Atlantic. Therefore, most of the significant TC related rain events in the Mid-Atlantic are a hybrid between a tropical system and an eastward advancing Maddox synoptic or frontal system.
However, based on the large-scale patterns it is clear that Connie remained independent of such a baroclinic system. Therefore, in this case, Connie was a relatively rare event. Only four primarily tropical events were among the top fifty 24-hour rainfall periods in our database.
The standardized anomalies of key fields often associated with heavy rainfall (Grumm and Hart 2001: Graham and Grumm 2010) were present here. This suggests that this event is another example on the value of standardized anomalies in identifying potential significant precipitation events. These concepts can be applied to numerical weather prediction output. This would assist in identifying potentially significant rainfall events.
One might have expected that Connie would have been one of the largest flood events on record, based on the expansive and persistent rainfall. However, despite the record rainfall from Connie, the flooding impact was relatively minor. This is proof that the magnitude of flood events is dependent on many variables other than rainfall such soil moisture, pre-event river levels, and in the cold season, snow cover.
This can be further illustrated by Hurricane Diane, which approached the region almost immediately after the remnants of Connie had dissipated. Although, Diane produced much less rainfall across the Mid-Atlantic than did Connie, there was significantly more flooding because the soil in the region was saturated and river levels were already at or above normal levels.
We would like to thank the NWS SCEP program. The precipitation typing project was part of the local SCEP training program.
Graham, Randall A., Richard H. Grumm, 2010: Utilizing Normalized Anomalies to Assess Synoptic-Scale Weather Events in the Western United States. Wea. Forecasting, 25, 428-445
Grumm, R.H. and R. Hart. 2001: Standardized Anomalies Applied to Significant Cold Season Weather Events: Preliminary Findings. Wea. and Fore., 16,736–754.
Hart, R. E., and R. H. Grumm, 2001: Using normalized climatological anomalies to rank synoptic scale events objectively. Mon. Wea. Rev., 129, 2426–2442.
Higgins, R. W., J. E. Janowiak, and Y.-P. Yao, 1996: A gridded hourly precipitation data base for the United States (1963-1993). NCEP/Climate Prediction Center Atlas 1, national Centers for Environmental Prediction, 46pp.
Jarvinen, B. R., C. J. Neumann, and M. A. S. Davis, 1984: A tropical cyclone data tape for the North Atlantic Basin, 1886-1983: Contents, limitations, and uses. NOAA Technical Memorandum NWS NHC 22, Coral Gables Florida, 21 pp.
Maddox, R. A., C. F. Chappell, L. R. Hoxit, 1979: Synoptic and Meso-α Scale aspects of Flash Flood Events1. Bull. Amer. Meteor. Soc., 60, 115–123.
Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40- Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77,437–471.