Skip Navigation

March-April 2013

Print
Email
ResizeResize Text: Original Large XLarge

Weather Queries

I believe that Hurricane Sandy and a severe East Coast storm around Thanksgiving in 1950 had many similarities, both in track and intensity. Can you comment on this?

Tom Rogers

Mechanicsburg, Pennsylvania

The two storms, one called the Thanksgiving Day Storm or the Great Appalachian Storm of November 25-27, 1950, and the other called Hurricane or Superstorm Sandy of October 28-31, 2012, have several things in common: They both caused widespread havoc in the Northeast United States, including coastal flooding along the mid-Atlantic coast; their central pressures dropped rapidly to very low values; their movement was anomalous (toward the west or northwest during their most energetic phases); and a blocking high over Southeast Canada stopped their eastward progress.

The two storms differed markedly in other ways. The Thanksgiving Storm of 1950 developed in response to a very strong short wave in the jet stream moving southeast from Canada, when it passed over a stalled front in the southern Appalachian Mountains. Once this storm developed, it drew record-setting cold air into its circulation and caused widespread, excessive snowfall in the Ohio Valley and central Appalachian Mountains. By contrast, Superstorm Sandy originated as a hurricane in the Caribbean Sea and moved north, retaining tropical characteristics until it made landfall near Atlantic City, New Jersey. In the 24 hours before landfall, Sandy interacted with a strong trough, which was oriented northwest-southeast and dug southeastward toward the coast of the Carolinas. This was undoubtedly instrumental in maintaining Sandy's exceptionally low pressure until it made landfall.

Storm Tracks and Central Pressures

Figure 1 shows the tracks of the two storms. Sandy (blue track) became a hurricane south of Jamaica (just off the map) on October 24. The blue dots show 12-hour positions at 00Z and 12Z. (“Z” is shorthand for Greenwich Mean Time, which is five hours ahead of Eastern Standard Time.) The number following the hour gives the day of the month, which, for Sandy was October 2012. Sandy followed a curving path, paralleling the East Coast until it reached the latitude of Virginia, at which point it hooked westward toward the New Jersey coast. The central pressures for Sandy are listed in millibars (mb) at each position (courtesy of Chris Landsea at NOAA's National Hurricane Center). One remarkable aspect of Sandy is that its central pressure deepened almost to the time of landfall. The 946-mb (27.93-inch) pressure recorded at landfall ties for the lowest pressure ever recorded along the east coast north of Cape Hatteras, North Carolina. Only the Great Long Island (New York) Hurricane of September 21, 1938, registered a pressure as low as Sandy's. All-time record-low pressures were observed in Atlantic City and Trenton, New Jersey; Philadelphia and Harrisburg, Pennsylvania; and Baltimore, Maryland. Typical of hurricanes, Sandy weakened rapidly after moving onshore and resumed a northerly path into Canada after reaching as far west as South-Central Pennsylvania.

Caption: Figure 1. Tracks of the Thanksgiving Storm of 1950 (red) and Superstorm Sandy (blue). Blue dots show 12-hour positions of Hurricane Sandy. Times and dates are given as two digit numbers. For example, 12Z/27 means 1200 Greenwich Mean Time on October 27, 2012. Central pressure in millibars appears to the right of the date and time. Red squares mark six-hour positions of the Thanksgiving 1950 storm. The same time/date/pressure convention is used for Superstorm Sandy. For example, 1230Z/24 means 1230 Greenwich Mean Time on November 24, 1950.

The Thanksgiving Storm of 1950 (red track) began as a weak wave along a stationary surface front draped across the southern Appalachians. When a pool of very cold air aloft carried by a vigorous trough overtook this front from the northwest, surface pressure began falling rapidly around 00Z on November 25. The red squares illustrate six-hour positions of the low-pressure center. The Thanksgiving storm split into two distinct centers between 1230 and 1830Z on November 25. Both centers had minimum pressures of 987 mb, but the eastern one in Pennsylvania quickly filled, while the western one in northeast Ohio began moving southwestward and deepened further to a minimum of 978 mb in Central Ohio (in between the red squares shown on the map). A pressure of 983.7 mb registered at Dayton, Ohio, set a new November record. After stalling in Ohio for nearly 18 hours and slowly weakening, the storm looped back to the northeast and entered Ontario.

Blocking Highs, Strong Easterlies, and Coastal Flooding

At the same time that Hurricane Sandy was making landfall on the New Jersey coast with a central pressure of 946 mb, a surface high of at least 1030 mb lay over Labrador. The strongest pressure gradient, and the strongest surface winds between these two centers, existed from the New Jersey coast to Southern Connecticut. Gale-force winds extended for 945 miles across Sandy—a record for Atlantic hurricanes. Consequently, the over-water fetch of strong easterly and northeasterly winds that pounded the shoreline from New Jersey to Connecticut was remarkable. Wind gusts from 75 to 96 mph raked both northern and southern shorelines of Long Island Sound. These drove a storm surge west southwest down the narrowing confines of Long Island Sound and into the East River of Manhattan. Winds of similar strength drove the storm surge into Lower Bay and The Narrows south of Manhattan. Not only did this occur at high tide, but also during a full moon, which makes normal tides even higher. The storm surge at The Battery in lower Manhattan reached 13.9 feet—nine feet above normal high tide and more than three feet above the previous record high-water mark set during a major hurricane in 1821.

Sandy Hook, New Jersey, about 15 miles south of New York City, recorded its highest water level on record, exceeding the previous record by 3.2 feet, which was set during Hurricane Donna in 1960. As Sandy came onshore, strong southeast winds pushed water up the Delaware Bay to Philadelphia, Pennsylvania, where the water level reached 10.6 feet—5.4 feet above astronomical high tide, and also besting the previous record set during Hurricane Donna. The record or near-record storm surge along the New Jersey coast and around New York City caused damage of at least $60-80 billion, second only to that of Hurricane Katrina in 2005.

As Figure 1 shows, the center of the November 1950 storm never quite reached open water. Nevertheless, strong east to southeast winds north of the center did cause coastal flooding. The surface pressure gradient was extreme along the mid-Atlantic and Southern New England coast. At 1830Z on November 25, 1950, as the pressure in Southeast Pennsylvania fell to 987 mb, the pressure in Southern Labrador was 1045 mb. Peak wind gusts were 110 mph at Concord, New Hampshire; 108 mph at Newark, New Jersey; 100 mph at Hartford, Connecticut; and 94 mph in New York City, exceeding values registered during Sandy. However, the over-water fetch of these winds was not nearly so great, nor did the strong onshore winds last nearly as long as with Hurricane Sandy, and so storm surge was not as high. A footnote: Both storms inundated LaGuardia Airport.

The anomalous movement of both storms toward the west was caused by blocking high-pressure systems over Southeast Canada, which refused to budge, thus preventing the normal eastward progression of these storms. Figure 2 illustrates the 500-mb flow pattern at 0000Z on October 30, 2012, the time that Sandy made landfall. Note the closed anticyclone (contour value 584) over Newfoundland and the 500-mb ridge at 60°W longitude. Temporarily, at least, Sandy had no place to go but westward.

Caption: Figure 2. Height of the 500-mb pressure surface at 0000Z, October 30, 2012. Contour values are in tens of meters. Thus 540 means that the 500-mb pressure surface is 5,400 meters above sea level. The wind flow tends to follow the contours. Note the blocking high over Newfoundland and the closed low over Delaware, which is almost directly above the surface low at this time.

Figure 3 is a historical 500-mb map for 0300Z on November 26, 1950. At the time, 500-mb contours were drawn at 200-foot intervals instead of 60-meter intervals as in Figure 2, but there is strong similarity between the two flow patterns. A sharp ridge is over the Canadian Maritime Provinces with a 1045-mb surface high, previously mentioned, nearly underneath it. The 500-mb ridge extends south-southeast over the Atlantic Ocean near 60°W longitude. The closed cyclone over West Virginia is about to stall (see again the track in Figure 1), as it has no place to go. The upper-air balloon sounding system was still fairly new in 1950. The storm itself prevented the release of many balloons at 0300Z. In addition, there were few sounding locations in Southeast Canada, and so the shapes and positions of the contours must be considered as approximate.

Caption: Figure 3. Height of the 500-mb pressure surface at 0300Z, November 26, 1950. Contour values are in hundreds of feet, as was customary at the time. Thus, 178 means that the 500-mb pressure surface is 17,800 feet above sea level. Note the sharp ridge over the Maritime Provinces and the deep closed low over West Virginia. Dashed lines are 500-mb temperature contours at 5°C intervals. Short, bold arrows indicate track of 500-mb low. Winds are plotted at sounding sites.

Precipitation, Liquid and Solid

Some of the heaviest precipitation associated with Sandy fell in the Tidewater area of Virginia and Maryland, which includes Washington, D.C., and Baltimore, Maryland. This area received from five to 12 inches of rain from October 29-November 1, 2012. Because the weather had been dry before Sandy, no serious flooding occurred. Of greater interest are the heavy snows, from one to nearly three feet, that fell over higher elevations of the Appalachians from Northeast Tennessee and Western North Carolina northeastward along the Virginia/West Virginia border to far Western Maryland. Snowfall in association with a landfalling hurricane is unprecedented, as far as I know. Some lower elevations in Ohio received their first snow of the season.

The Thanksgiving Storm of 1950 brought two to six inches of precipitation, most of it rain, in a broad swath stretching from Northern Virginia across most of Pennsylvania, all of New Jersey, Southeast New York, and all the way to Central New England. Yet this storm was most notable for the very heavy snow it brought to parts of Ohio, Pennsylvania, and West Virginia, along with record-breaking low temperatures. An upper-air low carrying unseasonably cold air plunged southeast from Lake Michigan to South Carolina in just 24 hours (midday November 24 to midday November 25), causing the rapid deepening of the surface storm described earlier. As this happened, bitterly cold air swept in behind the surface low, wrapping around its south side and causing heavy snowfall. Relatively warm air swept in from the Atlantic Ocean on the north side of the circulation. At one point, it was raining in Buffalo, New York, with a temperature near 50°F, while at the same time in Western Pennsylvania, just 200 miles south, it was snowing hard with a temperature in the single digits.

Figure 4 gives the snowfall totals for the Thanksgiving Storm. Eastern Ohio, Western Pennsylvania, and much of West Virginia had over 18 inches of snow. Parts of West Virginia were buried in more than 40 inches. Parkersburg, West Virginia, with a snowfall record dating back to 1888, received 34.4 inches of snow from November 24-28, more than doubling the previous November record and, in fact, breaking the record for any month by more than five inches.

Caption: Figure 4. Snowfall (in inches) recorded from November 22-28 as a result of the Thanksgiving Day Storm of 1950. Colored shading represents increments of ten inches (light blue: 10-20 inches, light purple: 20-30 inches, red: 30-40 inches, and yellow: above 40 inches).

Table 1 lists some record minimum temperatures for November that were set during the 1950 storm. Note that a number of November records were shattered by 5°F or more at stations that had a long period of record.

Table: Table 1. Record Minimum November Temperatures (°F) Set During the Thanksgiving Storm of 1950.

Station

November 1950 minimum

Previous November record minimum

Years of record

ALABAMA

   

Birmingham

5

14

54

Mobile

22

24

78

Montgomery

13

18

79

FLORIDA

Apalachicola

24

28

29

Jacksonville

23

25

79

Pensacola

22

25

70

GEORGIA

   

Atlanta

3

14

72

Augusta

11

21

78

Savannah

15

22

79

KENTUCKY

   

Louisville

−1

1

78

NORTH CAROLINA

   

Asheville

−5

4

47

Wilmington

16

20

80

SOUTH CAROLINA

   

Charleston

17

23

79

Greenville

11

14

65

TENNESSEE

   

Chattanooga

4

11

72

Knoxville

5

8

79

Memphis

9

16

78

Nashville

−1

8

78

Credit: American Meteorological Society. This is Table 1 on p. 208 in the article entitled

“The Destructive Storm of November 25–27, 1950” by Clarence D. Smith, Jr., Monthly Weather Review, November 1950.




Weatherwise Contributing Editor THOMAS SCHLATTER is a retired meteorologist and volunteer at NOAA's Earth System Research Laboratory in Boulder, Colorado. Submit queries to the author at weatherqueries@gmail.com, or by mail in care of Weatherwise, Taylor & Francis, 325 Chestnut St., Suite 800, Philadelphia, PA 19106.



In this Issue

On this Topic

© 2010 Taylor & Francis Group · 530 Walnut Street, Suite 850, Philadelphia, PA · 19106 · heldref@taylorandfrancis.com