1. Overview of the season
The 2001 season for the eastern North Pacific basin (from 140°W eastward and from the equator northward) began a little early on 25 May (the median starting date is 31 May) with Adolph. It became the strongest hurricane and the only category 4 hurricane on the Saffir–Simpson hurricane scale (SSHS; Simpson 1974) on record in this basin for May since reliable records began in the mid-1960s. In terms of numbers of tropical storms (maximum 1-min surface wind speeds of 34–63 kt) and hurricanes (maximum 1-min surface wind speeds of 64 kt or higher), the season was near normal. There were 15 named storms of which 8 became hurricanes. In addition, there were two tropical depressions that failed to acquire tropical storm strength. For comparison, the 1966–2000 average is 15 named storms and nine hurricanes. There were only two “major” hurricanes (category 3 or higher). On average, four tropical cyclones reach such status. Two to three tropical storms or hurricanes hit Mexico every year. However, only one tropical cyclone, Juliette, hit the coast of Mexico in 2001. This hurricane reached category 4 intensity, but weakened to category 1 by the time it was near southern Baja California, Mexico. Juliette brought hurricane force winds to the Cabo San Lucas area, but weakened to a tropical storm by the time its center crossed the coast of the Baja California peninsula. It was responsible for two deaths. Adolph, Dalila, Ivo, and Lorena came close enough to the coast to require the issuance of watches or warnings.
Tropical wave records kept since 1967 indicate that most eastern North Pacific tropical cyclones are associated with westward-moving synoptic-scale disturbances, or tropical waves (Avila and Guiney 2000). In general, these waves amplify and trigger tropical cyclones when they reach this basin, where climatologically favorable conditions for tropical cyclogenesis such as low-level cyclonic vorticity, high moisture, and low shear prevail (Gray 1968). This year, the development of all tropical cyclones was associated with a tropical wave, or with the remnants of an Atlantic tropical cyclone. Figure 1 is a sequence of twice-per-day infrared satellite images in longitude versus time format (Hovmöller diagram) from 2 to 20 September of 2001. Note in the Hovmöller diagram the distinct areas of convection moving westward from Africa to the eastern North Pacific, associated with propagating synoptic-scale disturbances. A good example is the area of disturbed weather that crossed Central America on 8 September, and moved westward to the eastern North Pacific with increasing organization. The associated tropical wave was previously observed over Albrook, Panama, during 5–6 September as depicted in Figs. 2a and 2b, which are vertical time sections of wind and relative humidity for that station for September 2001. Note the cyclonic wind shift and the significant increase in moisture to above the 400-mb level associated with the passage of the wave. The value of the deviation from the mean for that period of the meridional component of the wind (υ′) reached −9 kt near 650 mb just after the wave axis moved by the station, implying that the wave had a large amplitude. This wave eventually triggered the development of Tropical Storm Ivo. In these figures, there are two other disturbances that passed by Albrook on 15 and 19 September, which led to the formation of Hurricanes Kiko and Juliette, respectively.
2. Tropical storm and hurricane summaries
The National Hurricane Center (NHC) produces a “best track” database for these tropical cyclones (TCs), which specifies center position, the maximum 1-min surface (10 m) wind speed, and the minimum sea level pressure. These parameters are determined at 6-hourly intervals after each cyclone has ended. Aerial reconnaissance for this basin, either by the United States Air Force Reserve Command or by the National Oceanic and Atmospheric Administration (NOAA), is sometimes operationally available when a TC is a threat to land, such as in the cases of Hurricane Juliette and Tropical Storm Lorena of 2001. The primary data source, however, is imagery from the Geostationary Operational Environmental Satellites (GOES) and polar-orbiting weather satellite imagery, from which position and intensity estimates are obtained using the Dvorak (1984) technique. A recent study by Brown and Franklin (2002) indicates that nearly half of Dvorak satellite-based intensity estimates fall within 7 kt of reconnaissance-based best track values, and only in 10% of the cases are the errors 20 kt or higher. Other valuable remotely sensed data sources are multichannel passive microwave imagery (Hawkins et al. 2001) and scatterometer data (Tsai et al. 2000). In addition, the reconstruction of the best tracks includes postanalysis of radar images and observations provided by the Servicio Meteorologico Nacional of Mexico as well as ship and buoy data from the area.
A summary of the 2001 named tropical cyclone statistics is included in Table 1 and their tracks are included in Fig. 3. The official track and intensity forecast errors are included in Tables 2 and 3. The individual tropical cyclone summaries in the next section are based on the NHC postanalysis. More detailed information on these systems may be found online (http://www.nhc.noaa.gov/pastall.html).
a. Hurricane Adolph, 25 May–1 June
A tropical wave that emerged from the coast of Africa on 7 May appears to have been the precursor disturbance of Hurricane Adolph. Surface analyses showed a weak low along the wave axis over Panama and Costa Rica beginning late on 18 May. The low eventually moved into the eastern North Pacific Ocean on 22 May with broadly distributed and disorganized convection. It was not until about 1800 UTC 25 May that a concentration of deep convection developed just south of the Gulf of Tehuantepec, and National Aeronautics and Space Administration's (NASA) Quick Scatterometer (QuikSCAT) satellite data suggested that the season's first tropical depression formed from this system about 215 n mi south-southwest of Acapulco, Mexico.
The tropical depression strengthened as it drifted to the east-northeast, and became Tropical Storm Adolph at 1800 UTC 26 May, when the cyclone was centered about 195 n mi south-southwest of Acapulco. As Adolph turned slowly northward toward the Mexican coast on 27 May, a ragged banding eye surrounded by deep convection became visible in satellite imagery. Adolph became a hurricane around 1800 UTC that day when it was located about 220 n mi south of Acapulco. On 28 May, Adolph came under the influence of a midtropospheric ridge to its north and turned slightly to the left. It was then located about 145 n mi southwest of Acapulco. Once hurricane strength had been attained, the pace of intensification increased dramatically. Based on satellite estimates, Adolph reached its peak intensity of 125 kt at 0000 UTC 29 May, making it the most intense (and only category 4) hurricane on record for the month of May in the eastern North Pacific Ocean. Figure 4 is a visible satellite image of Adolph near the time of its peak intensity. It is worth noting that during the early stages of the deepening period from 1534 UTC 27 May to 0048 UTC 28 May, Special Sensor Microwave Imager (SSM/I) satellite data indicated the eye of Hurricane Adolph had decreased by at least one-half its original diameter of about 20 n mi. The significant reduction in the size of the eye was not apparent in conventional visible and infrared satellite data.
Shortly after reaching its peak intensity, Adolph turned more to the west. A slow weakening trend began with some oscillations in convective intensity and eye definition. By 0000 UTC 1 June, Adolph weakened below hurricane strength. Thereafter, convection continued to diminish and the system became a tropical depression by 1800 UTC that same day. No further deep convection was noted after that, and the tropical cyclone became a nonconvective area of low pressure about 400 n mi south-southwest of Cabo San Lucas, Mexico, by 0000 UTC 2 June. A swirl of low clouds persisted in this area for several days thereafter. There were no reports of damage or casualties associated with Hurricane Adolph.
b. Tropical Storm Barbara, 20–26 June
The origin of Barbara can be traced to a tropical wave that moved westward across the coast of Africa on 1 June. The wave moved into the Caribbean Sea on 7 June, and into the eastern North Pacific on 10–11 June with little signs of development. The wave continued westward to near 120°W by 18 June, where the shower activity increased. Further slow organization over the next 2 days resulted in the formation of a tropical depression at 0000 UTC 20 June, about 1150 n mi southwest of Cabo San Lucas. As the cyclone formed, it began a west-northwestward motion that would persist through its lifetime.
The depression intensified into Tropical Storm Barbara later on 20 June, then reached a peak intensity of 50 kt on 21 June. Thereafter, increasing southwesterly vertical shear caused weakening. Barbara became a depression as it crossed 140°W and passed into the central Pacific hurricane basin. The depression passed north of the Hawaiian Islands on 25–26 June, and then weakened to an open wave to the northwest of Kauai on 26 June. The remnants of Barbara continued west-northwestward until they were absorbed by a frontal zone near the international date line on 30 June.
The only observation of tropical storm force winds was from the ship V2FA2 (name unavailable), which reported 47 kt winds 105 n mi northwest of the center at 0000 UTC 22 June. However, this observation is at odds with concurrent satellite-based intensity estimates and may not be valid. There were no reports of damage or casualties.
c. Tropical Storm Cosme, 13–15 July
Cosme developed from a tropical wave that crossed Central America and emerged into the eastern North Pacific basin on 6 July. Hovmöller diagrams of satellite images (not shown) over the tropical Atlantic prior to 6 July are not conclusive, but indicate that the responsible system emerged from the African coast on either 24 or 27 June.
The wave moved slowly westward from 6 to 10 July. On 10 July, the convective pattern began to show signs of organization about 350 n mi south of Acapulco. Over the next 2 days, the system moved generally west-northwestward as multiple low-level circulations developed within a broad area of low pressure. During this period, development was hindered by southerly shear from an upper-level trough to the west of the disturbance. On 12 July, an upper-tropospheric cyclone cut off southwest of the disturbance and the organization improved. By early on 13 July, a single low-level circulation center had become established and it is estimated that a tropical depression formed at 0600 UTC about 330 n mi southwest of Manzanillo, Mexico.
The depression moved west-northwestward at about 15 kt and quickly reached tropical storm strength by 1200 UTC 13 July, about 425 n mi south of Cabo San Lucas. Its peak intensity of 40 kt was reached late on that day. Thereafter, Cosme's development was hindered by easterly shear and by early on 14 July convection was limited and well removed from the center. Cosme weakened back to a tropical depression by 1800 UTC, when it was about 400 n mi southwest of Cabo San Lucas and became a nonconvective low center at 0600 UTC 15 July. The low then moved slowly westward until it dissipated on 18 July about 820 n mi west-southwest of Cabo San Lucas.
d. Hurricane Dalila, 21–28 July
Dalila moved along the Pacific coast of Mexico for several days, but its center remained 100 n mi offshore. Some flash flooding that damaged dozens of homes in the province of Chiapas is attributed to Dalila's rainfall.
The cyclone's origin was a tropical wave that moved westward from Africa into the eastern tropical Atlantic Ocean on 10 July. It crossed northern South America and Central America on 15–17 July, accompanied by vigorous thunderstorm activity, and then entered the Pacific basin on 18 July as an organized area of disturbed weather.
Early on 21 July, the system developed a low-level circulation and became a tropical depression about 250 n mi south of the Gulf of Tehuantepec. It moved west-northwestward and became Tropical Storm Dalila 12 h later. Dalila's track was rather steady toward the west-northwest at forward speeds of between 5 and 15 kt around a subtropical ridge.
With warm sea surface temperatures and minimal vertical shear, the winds reached an estimated 65 kt on 24 July, but for only a few hours. Thereafter Dalila weakened to tropical storm status and passed directly over Socorro Island on the next day. It remained at 55 kt until early on 27 July. By then, most of the associated deep convection dissipated as the storm moved over colder water. Dalila became a swirl of low clouds on 28 July, while located about 650 n mi west of the southern tip of Baja California.
Dalila is an interesting case to demonstrate the limitations of satellite-based intensity estimates. Dvorak intensity estimates on 24 July were as high as 77 kt. However, the maximum winds were adjusted to 65 kt during that period. These 77-kt estimates were based on nighttime infrared images that showed a cold central dense overcast (CDO) cloud pattern located over what was presumed to be an embedded center. In contrast, satellite microwave data showed a low-level center partially exposed to the northeast of the CDO, rather than embedded within the convection. This was an indication of a weaker tropical cyclone and was the basis for the intensity adjustment in the best track.
e. Tropical Storm Erick, 20–24 July
Erick appears to have formed from a poorly defined tropical wave that moved westward across the tropical Atlantic and reached the eastern North Pacific on 16 July. The thunderstorm activity associated with the wave increased on 18 July when the disturbance was centered about 700 n mi south of the southern tip of Baja California. It was not until 1800 UTC 20 July that a well-defined center of circulation formed and satellite images and data from QuikSCAT supported tropical depression status. Moving on a general west-northwest track, the system became a tropical storm and reached maximum winds of 35 kt at 1200 UTC 22 July. It then moved over relatively cool waters and weakened as the deep convection quickly vanished. By 0000 UTC 24 July, it was a nonconvective and dissipating swirl of low clouds, although some showers redeveloped intermittently.
f. Hurricane Flossie, 26 August–2 September
Flossie originated from a tropical wave that moved across the west coast of Africa on 11 August and produced Tropical Storm Chantal over the tropical Atlantic. After Chantal made landfall near Chetumal, Mexico, the southern portion of the associated disturbance crossed Central America and emerged over the Pacific Ocean south of Guatemala on 21 August. For the next several days, the area of cloudiness and showers moved westward and west-northwestward on a track parallel to the southwest coast of Mexico, with little change in organization. On 25 August, outer convective bands associated with the disturbance moved onshore near Manzanillo, and produced wind gusts to tropical storm force. However, the convection was still too disorganized for the system to be classified as a tropical cyclone.
As the disturbance gradually moved away from the coastal mountains of Mexico, a low-level center of circulation became better defined and deep convection consolidated closer to this center. It is estimated that a tropical depression formed about 235 n mi south-southeast of the southern tip of Baja California at 0600 UTC 26 August. The depression moved westward, and by 1800 UTC on the same day, it strengthened into Tropical Storm Flossie. The storm continued westward, passing north of Socorro Island. Flossie strengthened into a hurricane by 1800 UTC 27 August. The intensification process leveled off, however, when Flossie experienced southwesterly shear associated with an upper-level low located to its west. Over the next 24 h, Flossie turned sharply southward and made a small cyclonic loop about 200 n mi west of Socorro Island. The upper-level low moved away from Flossie, and the hurricane began to intensify again. Flossie reached a maximum strength of 90 kt at 1800 UTC 29 August. As the cyclone moved northwestward, a slow weakening trend began and the system dropped below hurricane strength around 0600 UTC 31 August, when it was located about 470 n mi west of the southern tip of Baja California. Moving over cooler waters, the cyclone weakened to a depression by 1200 UTC 1 September. Flossie degenerated into a nonconvective low later that day about 200 n mi west of Punta Eugenia in western Baja California.
The remnant low-level circulation drifted slowly westward over the eastern North Pacific Ocean as a swirl of low clouds for several more days. However, the mid- to upper-level circulation moved rapidly northeastward across northern Baja California before dissipating over the desert region of the southwestern United States. There were no direct deaths or damage caused by Flossie while it was a tropical cyclone. However, there were deaths due to lightning associated with remnant moisture from Flossie that helped to trigger strong thunderstorms, deadly lightning, and flash floods across portions of southern California. Four people were struck by lightning in the San Diego and San Bernardino Mountains on 2 and 3 September, and two of the four victims died. In addition, more than 2 in. of rain fell in an hour and caused flash flooding in San Diego and Riverside Counties on 2–3 September. Strong winds also knocked a tree onto a house. Total damage from Flossie's remnants was estimated at $35,000.
g. Hurricane Gil, 4–9 September
A tropical wave moved westward across the African coast on 14–15 August. The system showed signs of organization as it approached the Lesser Antilles on 21 August. The northern portion of the wave spawned Tropical Storm Dean on 22 August and the southern one continued westward, crossing Central America into the eastern North Pacific on 24 August. Little development occurred for the next several days as the wave continued westward. Convection associated with the wave increased on 30 August while a large-scale low-level cyclonic turning developed over the tropical eastern North Pacific. The system gradually became better organized over the next few days, and a tropical depression formed around 0600 UTC 4 September, about 850 n mi southwest of Cabo San Lucas. The cyclone strengthened and became Tropical Storm Gil 6 h later. Meanwhile, the tropical depression that was to become Henriette formed about 765 n mi to the east of Gil.
Gil moved westward with its speed varying from 3 to 8 kt from 4 to 6 September. Steady strengthening occurred during this period, and Gil became a hurricane early on 6 September. It reached its peak intensity of 85 kt by 1800 UTC that day when an eye was visible in both conventional and microwave satellite imagery. At that time the stronger Gil was located 465 n mi southwest of the larger Henriette (Fig. 5). Gil turned northwestward early on 7 September. Later that day, it accelerated northward as Henriette began to pass to the north and the two cyclones began interacting in earnest. Gil began to weaken as it encountered shear caused by the upper-level outflow from Henriette. By 0000 UTC on 8 September, Henriette was passing 330 n mi north of Gil and the two cyclones began to rotate around each other. This caused Gil to move at 20–25 kt on a north-northwestward track that changed to west by 0000 UTC on 9 September and then weakened steadily.
While Henriette's circulation dissipated shortly after 1200 UTC on 8 September about 210 n mi west of Gil, the remnant low- to midlevel vorticity center continued to move around the south and east sides of Gil. This caused Gil to turn southwestward early the next day. Once Henriette's remnants were fully absorbed, Gil slowed from a 20–25-kt motion to a westward drift in a few hours. Associated convection totally dissipated during the merger and did not return afterward. This resulted in Gil becoming a nonconvective low late on 9 September about 1000 n mi east of the Hawaiian Islands. The remnant low cloud swirl moved generally west-northwestward to northwestward for several days, finally being absorbed by a cold front well to the north of the Hawaiian Islands on 19 September.
The combined circulations of Gil and Henriette helped trigger a strong surge of southwesterly and southerly flow to the east and southeast of the cyclones. The ship Pacific Highway encountered this flow, reporting 40-kt winds, 1005.3-mb pressure, and 22-ft seas at 0000 UTC 7 September while about 205 n mi southeast of the center of Gil.
The merger of two tropical cyclones, or the absorption of one tropical cyclone by another, are uncommon events in the National Hurricane Center's area of responsibility. The last documented case of such an occurrence in the eastern North Pacific was when Tropical Storm (later to become Hurricane) Norbert absorbed Tropical Depression Eighteen-E in September 1990 (Avila 1991).
h. Tropical Storm Henriette, 4–8 September
A Hovmöller diagram of satellite images indicates that Henriette's precursor was a tropical wave that crossed Central America on 28–29 August. In a monsoonlike environment characterized by large-scale low-level cyclonic turning, the wave began showing signs of development a couple hundred miles south of Acapulco on 1 September, and by 2 September convection had become better organized. A QuikSCAT pass suggested that a closed circulation may have been already present by 1200 UTC 3 September; however, visible satellite images showed that the circulation center was exposed northeast of the poorly organized deep convection due to strong easterly or northeasterly shear. Little change in organization occurred during the next 24 h as the system moved west-northwestward a few hundred miles south of Mexico.
Early morning visible satellite images on 4 September revealed a partially exposed but well-defined low-level circulation. While deep convection was confined to the southwestern half of the circulation, the convection was close enough to the center for Dvorak satellite intensity estimates to increase to 25–30 kt. The system became a tropical depression at 1200 UTC 4 September, about 300 n mi west-southwest of Manzanillo, and also about 765 n mi east of the tropical depression that was to become Hurricane Gil.
A midlevel ridge to the north of the depression took the system initially on a west-northwesterly track at about 14 kt. Early on 5 September, as the depression's heading turned to the west, the separation between the circulation center and the deep convection lessened, and Dvorak satellite intensity estimates increased to 35 kt. A QuikSCAT pass at 1347 UTC showed a surge of southwesterly winds of 40–45 kt as far as 250 n mi south of the center. On the basis of the Dvorak estimates and the QuikSCAT data, it is estimated that the depression became Tropical Storm Henriette at 0600 UTC 5 September, about 350 n mi south-southwest of Cabo San Lucas. At this time, Tropical Storm Gil was located about 690 n mi to the west of Henriette.
Henriette slowly became better organized on 6 September. The convective pattern became more symmetric and the intensity increased to 50 kt. Meanwhile, Henriette turned to the northwest and accelerated to a forward speed of 15–17 kt as it began to feel the influence of Hurricane Gil, then located 465 n mi to the southwest (Fig. 5). Upper-level easterly flow, which was still evident over the cyclone early on 6 September, lessened and a more favorable outflow pattern began to develop. Convective banding near the center became better defined, and Henriette reached its peak intensity of 55 kt at 0000 UTC 7 September.
At the time of peak intensity, water temperatures under Henriette were about 25°C, but rapidly decreased as the cyclone moved northwestward. Within a few hours, the convective cloud tops warmed significantly. As the convection collapsed, Henriette began to weaken and turned west-northwestward around the periphery of Hurricane Gil. At this point Gil, which had been moving westward, turned sharply north-northwestward, and the distance between the two cyclones began to decrease rapidly. By 0000 UTC 8 September, Henriette was over 22°C water, devoid of deep convection, and moving westward at 20 kt about 330 n mi north of Hurricane Gil. Henriette turned to the southwest at a speed of about 22 kt, and by 1200 UTC 8 September had weakened to a tropical depression. Well entrained into the circulation of Gil, Henriette dissipated when it lost its own closed low-level circulation, as indicated by low-cloud trajectories, shortly after 1200 UTC on 8 September. At the time of dissipation, Henriette was located about 210 n mi west of Tropical Storm Gil.
The cloud pattern associated with the remnant vorticity of Henriette could be tracked for over 24 h after the tropical cyclone formally dissipated. Henriette's remnants completed a cyclonic loop of Gil, and could be tracked until about 1600 UTC 9 September, when they were centered about 185 n mi north-northeast of Gil.
i. Tropical Storm Ivo, 10–15 September
Ivo formed from a large tropical wave that moved across the African coast on 26 August. The wave was accompanied by numerous thunderstorms and a large cyclonic rotation at the low to midlevels of the atmosphere and when it entered the eastern Atlantic. On 28 August, the wave spawned a nondeveloping northward-moving vortex in the eastern Atlantic, but the wave's southern portion continued westward with very limited convective activity. Once the wave moved over the western Caribbean Sea during 5–6 September, the shower activity increased. The system continued moving slowly westward over Central America and the cloud pattern gradually became better organized as indicated in Fig. 2. On 9 September, satellite images showed a low- to midlevel circulation centered near Acapulco, and by the next day, a portion of the system moved over water. It then became a tropical depression about 100 n mi south-southwest of Acapulco at 1200 UTC 10 September.
Under moderate easterly shear, the center of the depression moved slowly westward and west-northwestward with its circulation hugging the southwest coast of Mexico. This proximity to land prompted watches and warnings. Satellite images and a 37-kt wind reported at 0600 UTC 11 September by the ship ZDEB2 located about 135 n mi from the center of the tropical cyclone indicated that system had become Tropical Storm Ivo. Thereafter, there was only slight strengthening and Ivo reached its maximum intensity of 45 kt at 1800 UTC 12 September. The tropical storm moved toward the northwest and then west over increasingly cooler waters, and gradually weakened. It became a low pressure system devoid of convection by 0000 UTC 15 September.
j. Hurricane Juliette, 21 September–3 October
Hurricane Juliette was a category 4 hurricane that brought 80-kt winds and heavy rain to southern Baja California. It also caused floods in the state of Sonora on mainland Mexico, and caused two deaths.
The tropical wave that produced Atlantic Tropical Depression Nine on 19 September over the western Caribbean was also the origin of Juliette. Tropical Depression Nine dissipated over Central America on 20 September but its remnants and associated tropical wave continued westward over the eastern North Pacific Ocean and became a depression at 0600 UTC on 21 September, about 90 n mi south of the coast of Guatemala. The cyclone moved on a track that was approximately parallel to the southwest coast of Mexico from 21 to 26 September. The center remained from 100 to 200 n mi offshore. Within an environment of weak vertical shear, the tropical cyclone strengthened and became a hurricane on 23 September. Juliette rapidly reached an intensity of 115 kt, with a “pinhole” eye appearing on 24 September and reached 125 kt and 923 mb on the next day. Whereas a number of hurricanes have had lower pressures estimated from satellite imagery, the aircraft-measured 923 mb on 25 September is the second-lowest measured sea level pressure on record in the eastern North Pacific Ocean. Only Hurricane Ava in June 1973 had a lower measured pressure, 915 mb. However, aircraft measurements in the eastern North Pacific are infrequent. Figure 6 shows a satellite image of Juliette at the time of peak intensity.
By 26 September, a strong trough west of the U.S. west coast began digging southeastward. Juliette gradually turned toward the north and began weakening. Moving very slowly northward, the center passed just west of Cabo San Lucas on the southern tip of Baja California on 28 September with maximum winds of 80 kt. These winds, heavy rain, and waves pounded the southern Baja California peninsula. Influenced by increasing vertical shear and progressively cooler sea surface temperatures, Juliette moved inland as a weakening 35-kt tropical storm on 30 September near San Carlos on the west coast of Baja California. The best track estimate of 35 kt at landfall is rather uncertain as the wind speeds were decreasing quickly. Juliette then moved over the waters of the central Gulf of California as a depression and finally dissipated on 3 October over the far northern Gulf of California. Its remnants spread into New Mexico, Arizona, and southern California. Figure 1b shows a continuous depression stage track across Baja California and the northern Gulf of California on 30 September; however, it is possible that the circulation dissipated over or near the southern Baja peninsula early on that day and another center formed in the northern Gulf of California.
The highest sustained wind observation from Baja California was 76 kt, with a gust to 94 kt at about 0000 UTC on 30 September from an automatic weather station at Cabo San Lucas. Rainfall totals include 136 mm on 27 September at Santiago on the extreme southern peninsula, 167 mm on 30 September at Empalme in the Sonora Province on the mainland, and 207 mm on 3 October at San Felipe in the extreme northern peninsula. An unconfirmed report was received of 300 mm at Cabo San Lucas.
Two deaths are attributed to Juliette. The Associated Press stated that a U.S. tourist surfing near the Baja California coast drowned in high seas on 27 September. The Mexican government news agency Notimex reported that a fisherman died near Acapulco when his small open boat capsized in high seas on 24 September. The Notimex news agency also reported that Juliette “clobbered” the tourist resort area of Cabo San Lucas, isolating it from the outside world for several days. The hurricane also caused flooding on mainland Mexico, driving more than 38 000 people from their homes in coastal areas of Sonora. Thunderstorm activity associated with the cyclone's remnants moved into southern California on 30 September, knocking down trees and power lines across the Coachella valley.
k. Hurricane Kiko, 21–25 September
A tropical wave that led to the formation of Atlantic Hurricane Felix over the eastern Atlantic on 7 September also appears to have produced Kiko. This wave moved westward at low latitudes, crossing northern South America on 13–14 September and Central America on 15 and 16 September as noted in Fig. 2. By 17 September, cloudiness and showers increased near the Gulf of Tehuantepec. The area of disturbed weather moved westward for the next few days, without much increase in organization. On 21 September, the system's cloud pattern became more consolidated, and curved bands of showers were evident. It is estimated that a tropical depression had formed by 1800 UTC that day, at which time it was centered about 550 n mi southwest of the southern tip of Baja California.
After forming, the system, which was located in an environment of easterly vertical shear, strengthened slowly and became a tropical strom by 1200 UTC 22 September. Kiko turned from a northwestward to a west-northwestward heading that day. Although some easterly shear continued to affect the system, very deep convection persisted near the center, and Kiko strengthened into a hurricane around 1200 UTC 23 September. A little later on 23 September, deep convection decreased in coverage and intensity and Kiko weakened back to a tropical storm. The system continued to lose intensity on 24 September at least in part due to the entrainment of more stable air at low levels. Kiko weakened to a tropical depression on the next day, by which time southwesterly shear also became prevalent. Later on 25 September, the cyclone degenerated into a westward-moving swirl of low clouds with little or no deep convection. Kiko's remnant low persisted and continued moving generally westward for several more days with intermittent, minor occurrences of deep convection within the circulation. It was finally absorbed into a frontal system to the northeast of the Hawaiian Islands on 1 October.
l. Tropical Storm Lorena, 2–4 October
The tropical wave that triggered Lorena moved across the west coast of Africa on 13 September. The poorly defined wave moved rapidly westward across the Atlantic for more than a week. There was little or no thunderstorm activity associated with the system until it moved across Central America on 27 September. Significant deep convection began to develop on 29–30 September when the wave was passing about 300 n mi south of Acapulco. Satellite intensity estimates indicate the system became a tropical depression at 0000 UTC 2 October, and by 1200 UTC it strengthened into Tropical Storm Lorena about 350 n mi south-southwest of Acapulco. Lorena moved steadily west-northwestward and gradually turned toward the northwest early on 3 October. The peak intensity of 50 kt occurred around 1200 UTC that day as Lorena took a more northerly track. It was then located about 180 n mi southwest of Manzanillo. Because Lorena was forecast to make landfall along the southwestern coast of Mexico near hurricane strength, a U.S. Air Force Reserve reconnaissance aircraft was in the storm between 1700 and 2300 UTC 3 October. The aircrew found that Lorena had already begun to weaken rapidly, based on 700-mb flight-level winds of only 29 kt. However, two dropwindsondes released about 100 n mi northeast of the center (outside of any convection) indicated wind speeds of 25–30 kt in a deep layer from the surface to 750 mb. These wind observations would suggest that minimal tropical storm force winds possibly existed closer to the cyclone center.
By 0000 UTC 4 October, Lorena's forward speed had decreased to 6–8 kt and strong upper-level southwesterly shear began to adversely affect the cyclone. Lorena weakened to a tropical depression at 1200 UTC and degenerated into a nonconvective low about 120 n mi southwest of Puerto Vallarta later on that day. The remnant low-level cloud circulation remained offshore and persisted for another day or so before dissipating just west of Cabo Corrientes, Mexico.
m. Tropical Storm Manuel, 10–18 October
Manuel formed from the remnants of Atlantic Hurricane Iris, which struck southern Belize as a category 4 hurricane (on the Saffir–Simpson hurricane scale) early on 9 October. By 1800 UTC, the core circulation of Iris had dissipated over the mountains of eastern Mexico, while new convection was developing a short distance away over the waters of the Pacific. This area became better organized over the next 18 h and became Tropical Depression Fifteen-E at 1200 UTC 10 October, about 175 n mi south-southeast of Acapulco. (Had the system remained a tropical cyclone throughout its passage over land, it would have retained the name Iris.)
The depression moved at 13–14 kt, first westward and then west-northwestward. An upper-level anticyclone centered over southern Mexico was producing some easterly shear in the environment of the depression, but when this shear diminished the system became Tropical Storm Manuel at 0600 UTC 11 October, about 200 n mi south-southwest of Zihuatanejo. An estimated peak intensity of 45 kt was first reached at 1800 UTC that day when the first well-defined banding features developed. However, they were short lived, deep convection diminished, and satellite microwave imagery early on 12 October suggested that the circulation was becoming elongated. Wind shear returned, this time from the northwest, and Manuel turned to a west-southwesterly track and slowed. By 1200 UTC 12 October, Manuel had weakened to a tropical depression.
Manuel remained a disorganized depression for the next 2.5 days. It continued moving to the west-southwest, but slowed to a drift as a midlevel ridge to the north of the cyclone gradually weakened. An upper-level trough dug southward to the west of Manuel early on 15 October, and Manuel began to move to the north-northwest. Convection redeveloped near the center and Manuel regained tropical storm strength at 0600 UTC 15 October, about 520 n mi south-southwest of Cabo San Lucas. Wind shear decreased and Manuel strengthened, reaching its peak intensity of 50 kt near 1200 UTC 16 October about 540 n mi southwest of Cabo San Lucas. By this point, water temperatures under the cyclone were decreasing and shear, this time from the southwest, was increasing. Manuel began to weaken and became a depression at 1800 UTC 17 October about 660 n mi west-southwest of Cabo San Lucas. It degenerated into a nonconvective low shortly after 0000 UTC 18 October. The remnant low moved slowly westward for a couple of days over cool waters before its circulation dissipated completely.
n. Hurricane Narda, 20–25 October
Narda developed from a westward-moving tropical wave that crossed Dakar, Senegal, around 3 October. The wave became convectively active after it crossed Central America when it produced a large burst of convection in the Bay of Campeche on 15 October. The southern portion of the wave continued westward over the Pacific waters south of Mexico, and under favorable upper-level winds, it began to acquire banding features with several centers of circulation. The system consolidated and developed one circulation center at 1200 UTC 20 October. It became a tropical depression about 1150 n mi southwest of Cabo San Lucas, Mexico. Moving on a west-northwestward track, it intensified and reached tropical storm status later that day. The cloud pattern continued to become better organized and visible satellite imagery intermittently showed an eye. It is estimated that Narda became a hurricane at 1800 UTC 21 October. Based on 3-h average objective Dvorak T numbers, Narda reached a peak intensity of 75 kt at 0000 UTC 22 October.
Thereafter, gradual weakening began and strong shear took a toll on Narda. The tropical cyclone became a tight swirl of low clouds with intermittent convection on 24 October, as it moved westward steered by the low-level flow and crossing 140°W over the central Pacific area of responsibility. It then continued westward as a tropical depression until dissipating into a nonconvective remnant low early on 25 October.
o. Hurricane Octave, 31 October–3 November 2001
Octave originated in the intertropical convergence zone and its development was likely initiated by a weak tropical wave that had moved westward across Central America on 22 October. Convection began to increase on 27 October, and a low-level circulation gradually developed. The system became a tropical depression at 0000 UTC 31 October while centered about 1000 n mi southwest of the southern tip of Baja California.
The cyclone was initially located to the south of a midlayer ridge, but a weakness soon developed in this ridge from a trough approaching from the west. This resulted in a general west-northwestward to northwestward track. Under light vertical shear, the depression gradually strengthened and developed a ragged eye that was visible for a few hours in satellite images. Octave's maximum winds reached an estimated 75 kt on 1 November. Vertical shear increased, and Octave weakened to a dissipating swirl of low clouds by 3 November about 1300 n mi west-southwest of the southern tip of Baja California.
3. Forecast evaluation
The NHC evaluates all official track and intensity forecasts of tropical storms and hurricanes by comparing them with the poststorm best track data. The track error is defined as the great circle distance between forecast and best track position of the tropical cyclone center, while the intensity error is the absolute difference between the forecasts and the best track 1-min wind speed. Tables 2 and 3 list the 2001 average official track and intensity forecast errors along with the previous 10-yr averages. Also included in these tables are the average Climatology and Persistence (CLIPER) and Statistical Hurricane Intensity Forecast (SHIFOR) model forecast errors for 2001 and the previous 10-yr average errors for these models. CLIPER (Neumann 1972) and SHIFOR (Jarvinen and Neumann 1979) are simple statistical (climatology and persistence) models for forecasting track and intensity, respectively, and serve as benchmarks for forecast skill. Table 2 shows that, in addition to being skillful, the official track forecasts have improved in comparison to the past 10 yr, especially at 48 and 72 h. It should also be pointed out that the CLIPER errors at 12–48 h were higher than the previous 10-yr average CLIPER errors, suggesting that the level of difficulty of the track forecasts was higher in 2001 than in the past 10 yr for these periods. Table 3 shows that the average official intensity forecast errors for 2001 are about the same as the average 2001 SHIFOR errors at 12–72 h, indicating little or no skill in the official intensity forecasts for this year. Thus, although the average official intensity forecast errors for 2001 are noticeably smaller than the longer-term averages, the mean SHIFOR errors for 2001 are also considerably smaller than their 10-yr average, indicating no real improvement with respect to the past decade.
REFERENCES
Avila, A. L., 1991: Eastern North Pacific hurricane season of 1990. Mon. Wea. Rev., 119 , 2034–2046.
Avila, A. L., and J. Guiney, 2000: Eastern North Pacific hurricane season of 1998. Mon. Wea. Rev., 128 , 2990–3000.
Brown, D. P., and J. L. Franklin, 2002: Accuracy of pressure–wind relationships and Dvorak satellite intensity estimates for tropical cyclones determined from recent reconnaissance-based “best track” data. Preprints, 25th Conf. on Hurricane and Tropical Meteorology, San Diego, CA, Amer. Meteor. Soc., 458–459.
Dvorak, V. F., 1984: Tropical cyclone intensity analysis using satellite data. NOAA Tech. Rep. NESDIS 11, NOAA, Washington, DC, 47 pp.
Gray, W. M., 1968: Global view of the origin of tropical disturbances and storms. Mon. Wea. Rev., 96 , 669–700.
Hawkins, J. D., T. F. Lee, J. Turk, C. Sampson, F. J. Kent, and K. Richardson, 2001: Real-time Internet distribution of satellite products for tropical cyclone reconnaissance. Bull. Amer. Meteor. Soc., 82 , 567–578.
Jarvinen, B. R., and C. J. Neumann, 1979: Statistical forecasts of tropical cyclone intensity for the North Atlantic basin. NOAA Tech. Memo. NWS NHC-10, NOAA, Washington, DC, 22 pp.
Neumann, C. B., 1972: An alternate to the HURRAN (hurricane analog) tropical cyclone forecast system. NOAA Tech. Memo. NWS SR-62, NOAA, Washington, DC, 24 pp.
Simpson, R. H., 1974: The hurricane disaster potential scale. Weatherwise, 27 (August) 169, 186.
Tsai, W-Y., M. Spender, C. Wu, C. Winn, and K. Kellogg, 2000: SeaWinds of QuickSCAT: Sensor description and mission overview. Proc. Geoscience and Remote Sensing Symp., Vol. 3, Honolulu, HI, IEEE, 1021–1023.
Time sequence of GOES infrared images (Hovmöller diagram) taken twice a day at 1145 and 2345 UTC from 8 to 20 Sep 2001. The latitude belt is roughly 5°–20°N
Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0249:ASNPHS>2.0.CO;2
(a) Vertical time section of the perturbation of the meridional wind (υ′; contour interval, 3 kt) at Albrook, Panama (station location 8.9°N, 79.6°W), from 1 to 30 Sep 2001. Here, υ′ is defined as υ minus the mean of the meridional wind for the same period. Positive and negative values are northerly and southerly components, respectively. Shaded areas denote positive values of the perturbation. Total perturbation wind (u′ and υ′) barbs plotted every 24 h (when available) according to standard meteorological convention with each full barb and half barb denoting 10 and 5 kt. (b) Same as in (a) but for the total wind. Dark-shaded areas denote areas of relative humidity higher than 80%
Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0249:ASNPHS>2.0.CO;2
(a), (b) Eastern North Pacific tropical storm and hurricane tracks for 2001.
Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0249:ASNPHS>2.0.CO;2
(Continued)
Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0249:ASNPHS>2.0.CO;2
NOAA-16 satellite image of Hurricane Adolph near the time of the cyclone's peak intensity. The multispectral false color image has been converted into a grayscale image
Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0249:ASNPHS>2.0.CO;2
GOES-10 satellite image of Gil and Henriette during their interaction at 1700 UTC 6 Sep 2001
Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0249:ASNPHS>2.0.CO;2
(left) GOES-10 image and (right) a view of Juliette captured by the Multi-angle Imaging Spectroradiometer (MISR) on 26 Sep. The MISR picture is provided by NASA and was taken during the Convection and Moisture Experiment-4 (CAMEX-4) hurricane study
Citation: Monthly Weather Review 131, 1; 10.1175/1520-0493(2003)131<0249:ASNPHS>2.0.CO;2
Eastern North Pacific basin hurricane season statistics for 2001
NHC eastern North Pacific basin average track forecast errors (n mi) for 2001
NHC eastern North Pacific basin average intensity forecast errors (kt) for 2001