Tropical Cyclone Report

Hurricane Olga

24 November - 4 December 2001

Lixion A. Avila
National Hurricane Center
17 December 2001

Tenacious late season Hurricane Olga roamed over the Atlantic Ocean for about a week.

a. Synoptic History

Hurricane Olga had a non-tropical origin and formed from the interaction of a cold front and a small area of disturbed weather in the north Atlantic. Five previous tropical cyclones and gale centers had already originated from a similar pattern in that area during this season. Olga was an uncommon event but not unique. There have been other late- or off-season tropical cyclones which moved southwestward across the Atlantic, such as Hurricane Lili during December of 1984. The long and winding track of Olga is displayed in Figure 1.

Surface observations and satellite imagery showed falling pressures and increasing weather in the area between Bermuda and the Leeward Islands as early as the 21st of November. A cold front reached the area on the 22nd, resulting in the formation of an extratropical low. The extratropical low gradually intensified and it is estimated that it transformed into a subtropical storm at 0000 UTC on the 24th, about 780 n mi east-southeast of Bermuda. At this time, satellite imagery showed the typical pattern of a subtropical cyclone with most of the weather within a comma-shaped cloud band located to the east of a low-level circulation center and a large area of gale force winds extending far to the north. During the next several hours, the thunderstorm activity markedly increased near an inner center of circulation which began to form, suggesting that the system was acquiring tropical characteristics. However, the wind field indicated that the system was a hybrid.

The shower activity continued to become more concentrated near the center and a post-analysis of satellite imagery indicates that by 1200 UTC 24 November, the inner portion of the system had already transformed into a tropical storm. In fact, there was a hint of an eye feature on visible satellite imagery at 1545 UTC on that day (Figure 2). However, one can argue that the system was not purely tropical, because it remained embedded within the much larger extratropical circulation.

The cyclone, now Tropical Storm Olga, moved slowly toward the northeast. As a high pressure ridge built to the north, Olga turned toward the west and west-southwest and gradually intensified. By then, satellite imagery showed well defined hooking bands surrounding a central dense overcast. Olga became a hurricane by 1200 UTC on the 26th and reached its peak intensity of 80 knots and a minimum pressure of 973 mb around 0600 UTC 27 November.

From the 26th to the 28th, Olga made a double cyclonic loop as it interacted with a large-scale deep-layer cyclonic circulation that was pretty much isolated from the main belt of westerlies. Once it completed the second loop on the 28th, Olga became steered southwestward by another strong area of high pressure to the northwest. It then encountered strong upper-level northerly and then westerly shear, displacing the convection from the center of circulation. Olga weakened to tropical depression status by 1200 UTC 30 November, but a temporary relaxation of the shear, allowed deep convection to redevelop near the center. Satellite imagery suggested that Olga regained tropical storm status at 0000 UTC 2 December. By then, Olga was moving northward and began to make another loop, this time anticyclonic.

Olga weakened as it was making this final loop and another high pressure ridge to the north steered the tropical cyclone west-southwestward. Olga gradually lost its associated deep convection and low-level circulation, and dissipated by 0000 UTC on the 5th about 600 n mi east of Nassau in the Bahamas. A vorticity maximum, embedded within a trough continued westward across the Bahamas, the Florida Straits, Cuba and the eastern Gulf of Mexico producing gusty winds and locally heavy showers.

b. Meteorological Statistics

Table 1 gives the best track positions and intensities of Olga at six-hourly intervals. Figure 1 shows a plot of this track. Figure 3 and Figure 4 depict the curves of maximum one-minute average (10 m above sea-level) wind speed and minimum sea-level pressure, respectively, as functions of time. Also plotted are the observations on which the curves are based. These consist of ship observations and data from satellite-based Dvorak- and Hebert-Poteat technique estimates by the Tropical Analysis and Forecast Branch (TAFB), Satellite Analysis Branch (SAB), and the U.S. Air Force Weather Agency (AFWA). The peak intensity associated with Olga was based on Dvorak objective T-numbers which were the highest around 0600 UTC 27 November.

There were several reports from vessels along the path of Olga and these are included in Table 2. The most significant observation came from a German sailing yatch Manana Tres which apparently went through the center of the cyclone and reported a minimum pressure of 989 mb at 0900 UTC 24 November. This was the basis for starting advisories. The ship Liberty Sun encountered the center of Olga during the 25th and the 26th, and captured the wind shift and the pressure change associated with the tropical cyclone as summarized in Table 2. Large waves associated with Olga spread onto the U.S. east coast, the Bahamas and most of the Caribbean Islands. In fact, some of the France's buoys reported waves in excess of 12 feet as far south as Guadeloupe.

c. Casualty and Damage Statistics

The only report of damage came from the Manana Tres which indicated "lots of damage" during its encounter with Olga.

d. Forecast and Warning Critique

Average official track errors for Olga were 63, 115, 141, 173, and 215 n mi for the 12, 24, 36, 48 and 72 h forecasts, respectively. With the exception of the 72 h forecasts, these errors were larger than last the 10-yr period as indicated in Table 3. Although the interaction of Olga with the large-scale deep-layer cyclonic circulation was considered in the official forecast, the tropical cyclone ended both loops and moved toward the west and west-southwest faster than anticipated. This probably caused large speed errors during that period. The NCEP Global model (AVNI) outperformed every other model and the official forecast. The absolute wind speed errors for Olga were 4, 7, 11, 14 and 12 kt for the 12, 24, 36, 48, and 72 h forecasts, respectively. For comparison the past 10-yr errors are 7, 11 14, 16 and 20 kt, for the same forecast periods.

Operationally, Olga was carried as a subtropical storm until 1500 UTC 26 November; while the best track shows the system becoming a tropical cyclone two days earlier - at 1200 UTC on the 24th. The lack of well-defined criteria separating subtropical from fully-tropical cyclones continues to be troublesome in these designations. In the initial advisory on the system on the 24th, it was noted that the low had a "small tropical-like center embedded within a huge non-tropical cyclone", and therefore was designated a subtropical cyclone (one that shares characteristics of tropical and non-tropical cyclones). The primary goal was to focus the marine community's attention on the large area of strong winds, which were largely unrelated to the developing inner core. Over the next two days the wind field gradually contracted in response to the persistent inner convection and the system was designated Olga on the 26th. An alternative interpretation, and the one taken in the construction of the post-storm best track, is that a small tropical cyclone had formed on the 24th with the development of the inner core, and that this tropical cyclones was embedded within a large-scale cyclonic environment.

Table 1: Best track for Hurricane Olga, 24 November - 4 December 2001.
Wind Speed
Lat. (°N)Lon. (°W)
23 / 060030.551.0100325extratropical
23 / 120030.251.0100025"
23 / 180029.750.799430"
24 / 000029.350.399050subtropical storm
24 / 060029.350.099050"
24 / 120029.549.898950tropical storm
24 / 180030.349.598950"
25 / 000031.350.698950"
25 / 060031.852.098950"
25 / 120031.753.598955"
25 / 180030.855.098760"
26 / 000030.355.397960"
26 / 060030.155.697960"
26 / 120030.655.997965hurricane
26 / 180031.156.097765"
27 / 000032.156.297575"
27 / 060032.657.097380"
27 / 120031.557.297480"
27 / 180032.056.097575"
28 / 000032.355.897575"
28 / 060032.655.797575"
28 / 120032.456.297775"
28 / 180031.457.097970"
29 / 000030.357.698765"
29 / 060029.358.799060tropical storm
29 / 120028.460.199260"
29 / 180027.361.899455"
30 / 000026.263.299845"
30 / 060025.364.7100135"
30 / 120024.566.2100330tropical depression
30 / 180023.767.1100430"
01 / 000023.467.7100430"
01 / 060023.168.0100530"
01 / 120022.868.3100430"
01 / 180023.068.6100430"
02 / 000023.468.9100335tropical storm
02 / 060024.368.9100335"
02 / 120025.569.2100140"
02 / 180026.469.5100140"
03 / 000027.069.7100435"
03 / 060027.469.9100535"
03 / 120027.869.9100535"
03 / 180027.769.6100535"
04 / 000027.769.1100630tropical depression
04 / 060027.868.2100630"
04 / 120027.868.0100630"
04 / 180027.067.4100630"
05 / 0000dissipated
27 / 060032.657.097380minimum pressure

Table 2: Selected ship observations of tropical storm or greater winds associated with Hurricane Olga, 24 November - 4 December 2001.
Ship Name or Call SignDate/Time (UTC)Lat. (°N)Lon. (°W)Wind dir/speed (deg/kt)Pressure (mb)
Dorothea Schulte24/000031.951.5010/38 1008.0 
Dorothea Schulte24/060032.350.3010/39 1008.0 
Irvin Primrose24/060034.252.2040/45 1012.5 
Manana Tres24/090029.551.0 989.0 
Liberty Sun24/120033.158.4350/36  
ELWZ724/120026.354.2320/43 1009.0 
Dorothea Schulte24/120032.649.1070/39 1005.0 
Irvin Primrose24/120034.253.8040/45 1013.5 
Dorothea Schulte24/180032.948.0070/39 1005.1 
Lykes Liberator25/090037.452.7050/47 1013.5 
Lykes Liberator25/120037.353.8040/44 1015.2 
Liberty Sun25/180030.856.1360/55 990.0 
Liberty Sun25/210030.155.8340/38 983.5 
Liberty Sun26/000029.755.5270/31 981.4 
Liberty Sun26/060030.154.5275/45 990.0 
Liberty Sun26/090030.453.9170/29 993.0 
Liberty Sun26/120030.753.4160/38 995.0 
Safmarine Infanta26/120024.158.2320/37 1004.0 
V7CR426/150032.455.8090/39 986.0 
V7CR426/180032.656.3090/39 985.5 
V7CR427/000033.357.8050/31 986.0 
Sabina29/180026.762.2340/34 1001.2 
City of Alberni29/180027.265.5020/35 1008.0 
Lykes Liberator25/120037.353.8040/44 1015.2 

Table 3: Table 3. Forecast evaluation (heterogeneous sample) for Olga, 24 November- 4 December, 2001. Forecast errors for tropical storm and hurricane stages (n mi) are followed by the number of forecasts in parentheses. Errors smaller than the NHC official forecast are shown in bold-face type.
Forecast TechniquePeriod (hours)
CLIP86 (28)195 (24)281 (20)406 (18)488 (18)
GFDI49 (25)89 (21)161 (17)302 (15)576 (14)
LBAR76 (28)162 (24)242 (20)352 (18)483 (18)
AVNI44 (25)67 (21)88 (17)134 (14)186 (12)
BAMD60 (28)119 (24)170 (20)232 (18)318 (18)
BAMM62 (28)111 (24)154 (20)199 (18)233 (17)
BAMS74 (28)128 (24)169 (20)225 (18)298 (18)
NGPI51 (26)69 (22)74 (18)116 (16)195 (14)
UKMI63(19)110 (16)119 (14)130 (13)159 (8)
GUNS50 (19)80 (16)105 (14)159(13)269(8)
GUNA47 (19)71 (16)92 (14)143 (13)219 (8)
NHC Official63 (27)115 (23)141 (19)173 (17)215 (17)
NHC Official (1991-2000 mean)44 (2049)82 (1835) 118 (1646) 151 (1475) 226 (1187)
*Output from these models was unavailable at time of forecast issuance.

Figure 1: Best track positions for Hurricane Olga, 24 November to 4 December, 2001. Dotted line with X marks the locations of the vorticity maximum associated with the remnants of Olga.

Figure 2: GOES visible satellite image showing Olga as a tropical storm at 1545 UTC 24 November, 2001. Note the eye feature trying to form.

Figure 3: Best track maximum sustained surface wind speed curve for Hurricane Olga, 24 November to 4 December, 2001, and the observations on which the best track curve is based.

Figure 4: Best track minimum central pressure for Hurricane Olga, 24 November to 4 December, 2001, and the observations on which the best track curve is based.


Last modified: 30-Jan-2002