45th Anniversary of Hurricane Debbie seeding

debbie1969

Hurricane Debbie’s eye as seen from ESSA DC-6

On August 18th and 20th, 1969 Hurricane Debbie was subjected to a seeding experiment as part of Project STORMFURY.  STORMFURY was a joint Navy and Environmental Science Services Administration (ESSA was the predecessor of NOAA) project to test the hypothesis that seeding hurricanes with silver iodide would disrupt their circulation and reduce their winds.  Two previous attempts to seed hurricanes (Esther in 1961 and Beuhlah in 1963) had been inconclusive.  The Debbie flights were the most extensive weather modification experiments carried out to date for the Project and the ones that yielded the most encouraging results.  Although the experiments were covered in the press, the public’s attention was concentrated at that time on Hurricane Camille, which had just made landfall in Mississippi.

USNwc-121air

U.S. Navy Super Connie flying into Hurricane Debbie.

Notice ‘sharkfin’ radome housing vertical scanning radar on top.

Debbie was a classic Cape Verde hurricane, forming from an African easterly wave that had become a tropical storm on August 15th while still 1200 nautical miles east of the Antilles.  It’s forecast track took it directly into the STORMFURY allowed target area to the east of the Bahamas, so the fleet of fifteen Navy, Air Force, and ESSA aircraft were put on notice by August 16th and deployed to Puerto Rico on the 17th.  The next day, Debbie was within the allowed area but at the maximum distance at which the experiments could still be carried out.  The first monitoring aircraft took off at 5 AM and the first seeding time (Tango) was set for 9 AM.  The two ESSA DC-6 airplanes flew racetrack patterns in and out of the eye of the storm to monitor any changes due to the seeding, while the Navy Super Connies flew around the storm using their superior vertical scanning radars to document the storm’s structure and the Air Force C-130 monitored high-level cloud particles.  Both the ESSA B-57 and Air Force WB-47 carried out high-altitude surveillance.  The seeding was done by Navy A-6 Intruder jets deploying canisters of burning silver iodide compound into the outer part of the hurricane’s eyewall.  The seeding continued at two-hour intervals until 3 PM, while the aircraft switched in and out of their assignments throughout the day.  The last monitoring plane landed at 11 PM that night.

DebbieSeedmap

Location of Hurricane Debbie during seeding experiments

 

The STORMFURY team analyzed the results throughout the next day and were very encouraged when they saw evidence of a weakening of the storm’s peak winds from 98 knots to 68 knots.  Debbie was forecast to still be within the STORMFURY target area the next day, so another round of seeding experiments was planned.  Upon returning to the hurricane on August 20th, the monitoring planes found that Debbie had regained strength with maximum sustained winds of 99 knots.  They also found that Debbie had formed a double eyewall during the 19th and that the outer eyewall had winds almost as strong as the inner. The Intruders seeded just outside of this inner eyewall, until the last round.  By then the inner eyewall had all but disappeared so the seeding took place just outside the outer eyewall.  The next day, Debbie had moved beyond reach of Puerto Rico and the day was spent analyzing the second round of seeding.  There was another reduction of strength noted, from 99 knots to 84 knots.  While less of a reduction than the previous experiment, many of the scientists involved were greatly encouraged.

Debbie eye Aug 20 1969

Hurricane Debbie’s double eyewall on August 20th as seen on radar.

ESSA touted the results of these experiments at an end-of-season press conference, but the STORMFURY scientists were careful to note that this was only one set of experiments and that many more would be needed in order to distinguish naturally caused changes in the storms from those caused by the seeding.  Unfortunately for the Project, only one other round of seedings were carried out in Ginger in 1971.  The Project remained funded until 1982, but further candidate storms eluded the researchers.  In the end, it was concluded that the changes in Debbie could well have been the result of an eyewall replacement cycle and that no positive conclusions could be made about the effect of the seeding.

Some papers by HRD scientists relating to Hurricane Debbie:

  • Bergman, K. H., and T. N. Carlson, 1975:  Objective analysis of aircraft data in tropical cyclones. Mon. Wea. Rev., 103, 431–444.
  • Black, P. G., and R. A. Anthes, 1971:  On the asymmetric structure of the tropical cyclone outflow layer.  J. Atmos. Sci., 28, 1348–1366.
  • Black, P. G., H. V. Senn, and C. L. Courtright, 1972:  Airborne radar observations of eye configuration changes, bright band distribution, and precipitation tilt curing the 1969 multiple seeding experiments in Hurricane Debbie.  Mon. Wea. Rev., 100, 208–217.
  • Dorst, N. M., 2007:  The National Hurricane Research Project: 50 Years of Research, Rough Rides, and Name Changes.  Bull. Amer. Meteor. Soc., 88, 1566–1588.
  • Gentry, R. C., 1970:  Hurricane Debbie modification experiments, August 1969.  Science, 168, 473-475.
  • Hawkins, H. F., 1971:  Comparison of results of the Hurricane Debbie (1969) modification experiments with those from Rosenthal’s numerical model simulation experiments.  Mon. Wea. Rev., 99, 427–434.
  • Lewis, B. M., and H. F. Hawkins, 1982:  Polygonal eye walls and rainbands in hurricanes.  Bull. Amer. Meteor. Soc., 63, 1294–1301.
  • Sheets, R. C., 1973: Analysis of hurricane data using the variational optimization approach with a dynamic constraint.  J. Appl. Meteor., 12, 963–976.
  • Sheets, R. C., 1973:  Analysis of Hurricane Debbie modification results using the variational optimization approach. Mon.  Wea. Rev., 101, 663–684.
  • Willoughby, H. E., D. P. Jorgensen, R. A. Black, and S. L. Rosenthal, 1985: Project STORMFURY, A Scientific Chronicle, 1962-1983, Bull. Amer. Meteor. Soc., 66, 505-514.

Doppler radar quick-looks from 2:00 AM P-3 flight into Tropical Storm Bertha, 5 August 2014

As Bertha weakened back to a tropical storm while passing southeast of Cape Hatteras, NC a NOAA P-3 mission collected airborne Doppler radar data to use in initializing and evaluating model guidance.  Included here you see images of the horizontal winds within 300 km of Bertha sampled from the tail Doppler radar on the P-3 aircraft during the early morning of 5 August 2014. These images are at three altitudes (1 km, 3 km, and 6 km) and are a composite of winds from the P-3 Doppler pattern around Bertha. Also plotted on each analysis are the locations of dropsondes deployed by the P-3 (plotted using standard station symbols). These analyses show that Bertha continued to weaken and it’s structure looked worse than the flight before with a very asymmetric distribution of precipitation at 1-and 3-km altitude, with the bulk of the precipitation east-southeast of the center. At 6-km altitude there was much less precipitation  surrounding the center with isolated areas east and well south of the center. There is indication of a circulation center at 1- and 3-km altitude, with a broad area of tropical storm force winds 50-60 km east-southeast of the circulation center, and a hint of a secondary wind maximum 170-180 km south-southeast of the center. At 6-km altitude it is very hard to discern a circulation center, with a trough embedded in southwesterly flow in the vicinity of the low-level centers indicative of the increasing southwesterly shear over the storm as it interacted with a frontal zone off the east coast of the United States.

All the Bertha radar composites at 0.5-km height resolution are available at http://www.aoml.noaa.gov/hrd/Storm_pages/bertha2014/radar.html

Doppler radar quick-looks from 2:00 PM P-3 flight into Hurricane Bertha, 4 August 2014

As Hurricane Bertha passed east of Cape Canaveral, FL a NOAA P-3 mission collected airborne Doppler radar data to use in initializing and evaluating model guidance.  Included here you see images of the horizontal winds within 300 km of Bertha sampled from the tail Doppler radar on the P-3 aircraft during the afternoon of 4 August 2014. These images are at three altitudes (1 km, 3 km, and 6 km) and are a composite of winds from the P-3 Doppler pattern around Bertha. Also plotted on each analysis are the locations of dropsondes deployed by the P-3 (plotted using standard station symbols). These analyses show that while Bertha had become a hurricane it’s structure looked worse than the flight before with a very asymmetric distribution of precipitation at all altitudes, with the bulk of the precipitation south and east of the center. There is indication of a circulation center at all altitudes, with winds of hurricane strength 25-30 km east-southeast of the circulation center, and a secondary wind maximum 80-90 km southeast of the center. From 1-6 km altitude the circulations center is tilted 10-15 km toward the east-southeast indicative of increasing westerly shear returning.

All the Bertha radar composites at 0.5-km height resolution are available at http://www.aoml.noaa.gov/hrd/Storm_pages/bertha2014/radar.html

Doppler radar quick-looks from 2:00 AM P-3 flight into Tropical Storm Bertha, 4 August 2014

As Tropical Storm Bertha strengthened as it passed east of the Bahama Islands a NOAA P-3 mission collected airborne Doppler radar data to use in initializing and evaluating model guidance.  Included here you see images of the horizontal winds within 300 km of Bertha sampled from the tail Doppler radar on the P-3 aircraft during the early morning of 4 August 2014. These images are at three altitudes (1 km, 3 km, and 6 km) and are a composite of winds from the P-3 Doppler pattern around Bertha. Also plotted on each analysis are the locations of dropsondes deployed by the P-3 (plotted using standard station symbols). These analyses show that Bertha had a very asymmetric distribution of precipitation at all altitudes, with the bulk of the precipitation surrounding the east semicircle around the center and extending toward the east. There is indication of a circulation center at all altitudes, with winds of hurricane strength 30-35 km east-northeast of the circulation center, and a secondary wind maximum 150 km east-northeast of the center. From 1-6 km altitude the circulations center is only slightly tilted (5-10 km) toward the east indicative of the storm intensifying as the westerly shear relaxed.

All the Bertha radar composites at 0.5-km height resolution are available at http://www.aoml.noaa.gov/hrd/Storm_pages/bertha2014/radar.html

Doppler radar quick-looks from 2:00 PM P-3 and G-IV flights into Tropical Storm Bertha, 3 August 2014

As Tropical Storm Bertha passed north of the Turks and Caicos Islands (the island coast line is visible as black line in the lower left of the G-IV images) a NOAA P-3 and a G-IV mission collected airborne Doppler radar data to use in initializing and evaluating model guidance.  Included here you see images of the horizontal winds within 300 km of Bertha sampled from the tail Doppler radar on the P-3 and within 450 km of Bertha from the G-IV aircraft during the afternoon of 3 August 2014. These images are at three altitudes (1 km, 3 km, and 6 km) and are a composite of winds from the P-3 and G-IV Doppler patterns around Bertha. Also plotted on each analysis are the locations of dropsondes deployed by the P-3 and G-IV (plotted using standard station symbols). These analyses show that Bertha had a very asymmetric distribution of precipitation at all altitudes, with the bulk of the precipitation surrounding the center and extending primarily toward the south and east. There is indication of a circulation center at all altitudes, with stronger winds 45-50 km east of the circulation center, and a secondary wind maximum 200 km east-northeast of the center. From 1-6 km altitude the circulations center is tilted 20-25 km toward the east indicative of moderate westerly shear.

All the Bertha radar composites at 0.5-km height resolution are available at http://www.aoml.noaa.gov/hrd/Storm_pages/bertha2014/radar.html

HRD Monthly Science Meeting of August 2014 and Debrief for missions into Tropical Storm/Hurricane Bertha – 14 August 2014

The August science meeting was used to discuss the results from the 6 P-3 and 1 G-IV missions into Tropical Storm/Hurricane Bertha. The agenda for the discussion was:

  • Missions Overview (Reasor)
  • Science Discussions
    • 20140801H1 Ferry/Ocean Survey (Uhlhorn)
    • 20140802H1 Ocean Survey/Center Fix (Uhlhorn)
    • 20140803H1 TDR (Rogers/Reasor)
    • 20140803N1 TDR (Dunion)
    • 20140804I1 TDR (Bucci)
    • 20140804H1 TDR (Uhlhorn)
    • 20140805I1 TDR (Bucci)
  • Field Program Issues

Slides from the debrief are available at:

ftp://ftp.aoml.noaa.gov/hrd/pub/blog/meetings/2014/science/HRD_SciMeet_20140814_bertha_debrief.pptx

Screen Shot 2014-08-15 at 5.56.26 PM

45th Anniversary of Hurricane Camille

hurricane-camille

On the night of August 17-18th, 1969, Hurricane Camille came howling into Waveland, MS with estimated sustained winds of 175 mph (280 km/hr) and a storm surge of 24 feet (7.3 m) to become the second strongest hurricane on record to make landfall in the United States (after the Labor Day Hurricane of 1935).  Camille also became an iconic storm that entered the public consciousness as the benchmark against which other disasters are compared.

  • The Storm    Camille can be traced as an easterly wave back to the west coast of Africa, but it gave no signs of development until it was west of Jamaica on Aug. 14th.  Having formed a circulation the depression moved northwestward toward the western tip of Cuba and intensified into a 110 mph (175 km/hr) hurricane as it passed over the island.  Entering the Gulf of Mexico, it began to further intensify but NHC had trouble gauging by how much.  Most of the Navy Hurricane Hunters had deployed to Puerto Rico for Project STORMFURY’s experiments seeding Hurricane Debbie.  The two aging Super Connies left on duty had mechanical problems and, in addition, couldn’t penetrate a hurricane over a certain strength since they entered the eye at sub-cloud level.  Thanks to the new infrared sensor on the Nimbus 3 satellite, meteorologists could monitor the storm at night, but NHC hurricane specialists and Washington, DC based satellite experts disagreed over the meaning of Camille’s contracting cloud shield viewed in IR the night of August 16-17th.  NHC thought it meant Camille was strengthening but the Satellite Service thought Camille was weakening.  Finally, NHC director Dr. Bob Simpson asked the Air Force to send a reconnaissance plane from California to the Gulf to penetrate the hurricane at high altitude.  They found the central pressure had dropped to an astounding 905 mb.  Simpson then knew he had to “ring the bell hard” in putting out warnings to the Gulf coast.  In addition, he released the output of an experimental surge model (SPLASH) which forecast a 20-foot storm surge.  Evacuations were ordered from Florida to Louisiana as the hurricane consistently tracked west of the forecast path.

   No anemometers survived in the area where Camille came ashore, but a barometer in Gulfport, MS measured a low pressure of 909 mb.  From that a maximum sustained wind speed of 190 mph (305 km/hr) was estimated.  The winds battered anything the 24-foot storm surge didn’t obliterate, and the Mississippi coast was devastated.  Camille degenerated as quickly as it had intensified as it moved inland, but it brought copious rains to the southeastern and mid-Atlantic United States as it tracked north then eastward, exiting to the ocean over Virginia.  In all Camille killed 259 people, the majority in Virginia due to flash flooding.  And the cost reached US$1.4 billion.

  • The Fallout    Immediately following Camille’s landfall, Simpson visited the Gulf coast surveying the damage.  At that time, Vice President Spiro Agnew was also touring the devastation and several Government officials, including Simpson, were asked to brief him on their actions during the storm.  After relating NHC’s efforts to warn people about the hurricane, Simpson mentioned the difficulties he’d had regarding reconnaissance due to the aging or inadequate Hurricane Hunter fleets.  A bureaucratic firestorm ensued with much finger pointing, and Simpson feared he might lose his job due to his being so outspoken.  He was not, and both the Departments of Defense and Commerce eventually allocated funding to upgrade their hurricane aircraft.  Simpson also struggled during Camille to adequately convey to emergency managers and first responders what sort of devastation they should prepare for.  So he turned to his friend Herb Saffir to adapt Herb’s wind storm scale to specific hurricane application.  The result was the Saffir-Simpson Hurricane Scale, adopted in 1971 and first used in public alerts in 1973.
  • The Reanalysis     As part of the HURDAT Reanalysis Project, Hurricane Camille was reconsidered in 2014.  Although most track changes were minor, Camille’s formation into a tropical depression was extended back 18 hours.  The major alterations were to the landfall values.  The 909 mb central pressure reading was found to have occurred not at the center of the hurricane eye, but peripheral to it.  So the value at landfall was lowered to 900 mb.  Using more modern pressure-wind relationships, the new central pressure yielded a new maximum sustained winds of 175 mph (280 km/hr).  This new value is more consistent with subsequent inland wind measurements and our present understanding of decay rates for hurricanes moving inland.  The researchers also think that Camille might have undergone an eyewall replacement cycle before coming ashore, a phenomenon not known at that time.
  • The Name     In the 1960s, Atlantic hurricane names consisted only of women’s names and the lists were rotated every four years.  In addition, names were only retired for ten years.  So when ‘Carla’ was retired in 1961 it was replaced on the 1965 list with ‘Carol’.  ‘Carol’ had been retired in 1954 when that storm devastated New England but was now eligible for reinstatement.  It was also brought back for the 1969 list, but scientists from the National Hurricane Research Laboratory (NHRL) asked the naming committee in January of that year to permanently retire Carol, Edna, and Hazel since papers were still being written about them. The committee agreed but needed a replacement ‘C’ name.  NHRL’s Dr. Banner Miller suggested the name of hurricane specialist John Hope’s daughter Camille.  Camille Hope was involved in an advanced science and math program in high school and the previous year had carried out a required independent research project.  Her father had asked Miller to mentor her in her investigation of hurricanes and long-term atmospheric trends.  Miller was impressed by her thoughtful work and so suggested her name for the list.  “We kept it quiet for many years,” Camille said in a recent phone interview.  “The names on the list weren’t supposed to be for a particular person.”  Stu Ostro, a Weather Channel colleague of John Hope’s, let the cat out of the bag five years ago.  ‘Camille’ was scheduled to be replaced by ‘Cindy’, but a new 10-year rotating scheme was adopted in 1971.
  • The Papers    Some scientific articles written by HRD personnel using data from Camille
    • Black, P. G. and R. A. Anthes, 1971:  On the asymmetric structure of the tropical cyclone outflow layer.  J. Atmos. Sci., 28, 1348-1366.
    • Powell, M. D., and T. A. Reinhold, 2007:  Tropical Cyclone Destructive Potential by Integrated Kinetic Energy.  Bull. Amer. Met. Soc., 88, 513-526.

Tenth Anniversary of Hurricane Charley

Hurr_Charley_radar

Hurricane Charley at landfall as seen by NWS Tampa radar

 

On Friday August 13, 2004, Hurricane Charley came roaring ashore at Port Charlotte, FL.  The hurricane had rapidly intensified prior to its landfall.  Also, since its track was running parallel to the southwest Florida coast, any slight variation in course would bring a large deviation of the expected landfall point.  This made Charley’s arrival something of an unpleasant surprise for many residents.

Charley had formed four days earlier south of Barbados.  It tracked south of Jamaica and struck the western tip of Cuba as a Category 3 hurricane early on the morning of August 13th.  It weakened slightly after encountering the island but after passing over the Dry Tortugas it began a rapid intensification and swerved northeastward as its winds ramped up from 110 mph (180 km/hr) to 145 mph (230 km/hr) in just three hours.  By the time Charley struck Cayo Costa, FL, its winds had peaked at 150 mph (240 km/hr).  Because it was a small hurricane, Charley cut a narrow swath of damage across Captiva Island, Punta Gorda, and Port Charlotte.  (It literally cut a narrow channel through Captiva.) Moving across Florida its strength rapidly diminished but still caused damage in Orlando and Kissimmee before moving out to sea at New Smyrna Beach.  It then went on to strike South Carolina as a Category 1 hurricane, bringing high winds and heavy rains to the Carolinas and Virginia before becoming absorbed into a front.  Charley was responsible for 15 direct deaths in Jamaica, Cuba, and the United States and caused over US$16 billion in damages.

charleytCone0812

Hurricane Charley’s forecast track at 11 PM on August 12th

There was considerable controversy after the storm.  Although the Florida landfall was within NHC’s cone of uncertainty, many people had concentrated on the ‘skinny black line’ at the center of the cone which depicted landfall in the Tampa region and were surprised by the hurricane’s turn to the northeast.  In addition, many of the high wind warnings issued for Orlando were not carried with the same alert headers as hurricane warnings and were either missed or misinterpreted by emergency managers, TV meteorologists, and the public.  Charley was the first of a record four hurricanes to strike Florida during the very active 2004 hurricane season.

Here are a list of scientific papers with HRD authors or coauthors resulting from Charley:

  • Aberson,S, D., 2008:  Large Forecast Degradations due to Synoptic Surveillance during the 2004 and 2005 Hurricane Seasons. Mon. Wea. Rev., 136, 3138–3150.
  • DiNapoli, S. M., M. A. Bourassa, and M. D. Powell, 2012:  Uncertainty and Intercalibration Analysis of H*Wind.  J. Atmos. Oceanic Technol., 29, 822–833.
  • Lonfat, M., R. Rogers, T. Marchok, and F. D. Marks Jr., 2007:  A Parametric Model for Predicting Hurricane Rainfall.  Mon. Wea. Rev., 135, 3086–3097.
  • Majumdar, S. J., S. D. Aberson, C. H. Bishop, R. Buizza, M. S. Peng, and C. A. Reynolds, 2006:  A Comparison of Adaptive Observing Guidance for Atlantic Tropical Cyclones.  Mon. Wea. Rev., 134, 2354–2372.
  • Marchok, T., R. Rogers, and R. Tuleya, 2007:  Validation Schemes for Tropical Cyclone Quantitative Precipitation Forecasts: Evaluation of Operational Models for U.S. Landfalling Cases.  Wea. Forecasting, 22, 726–746.
  • Powell, M. D., and T. A. Reinhold, 2007:  Tropical Cyclone Destructive Potential by Integrated Kinetic Energy.  Bull. Amer. Met. Soc., 88, 513-526.
  • Reynolds, C. A., M. S. Peng, S. J. Majumdar, S. D. Aberson, C. H. Bishop, and R. Buizza, 2007:  Interpretation of Adaptive Observing Guidance for Atlantic Tropical Cyclones.  Mon. Wea. Rev., 135, 4006–4029.
  • Zhu, P., J. A. Zhang, and F. J. Masters, 2010:  Wavelet Analyses of Turbulence in the Hurricane Surface Layer during Landfalls.  J. Atmos. Sci., 67, 3793–3805.

10th Anniversary of Typhoon Rananim

Typhoon_Rananim_2004

NASA MODIS satellite photo of Typhoon Rananim

On August 12, 2004 Typhoon Rananim made landfall in China’s Zheijang province.  It had formed a week earlier and had tracked between Taiwan and Okinawa, where it produced winds and heavy rains on those islands and caused one death on Taiwan.  Its landfall on the Chinese mainland was marked by storm surge and winds of 90 knots (180 m/s) destroying over 64,000 homes and causing wide-spread power outages.  Torrential rains then caused heavy flooding as the storm moved inland.  It was the strongest typhoon to hit this part of the coast since 1956 and killed over 160 people and resulted in nearly US$2.5 billion in damages.  Its name was retired and replaced with “Fanapi”.