SAN JOSE, COSTA RICA
13 TO 17 APRIL 2015
RA IV/HC-37/Doc. 4.2(4)
REVIEW OF THE PAST HURRICANE SEASON Reports of hurricanes, tropical storms, tropical disturbances and
related flooding during 2014 Report from Bermuda (Submitted by Bermuda)
Hurricane Fay, 11-12th October Bermuda Weather Service (BWS) first contacted the Bermuda Emergency Measures Organisation (EMO) on 8th October after speaking with the NHC, Miami regarding concern a low that had not yet been identified by the NHC as a probability of formation in their 5 day outlook. The next day NHC gave the trough a 20% probability of formation within 5 days. This information was conveyed to the public via BWS forecasts, discussion, as well as via our regular local radio broadcasts highlighting the fact that we were monitoring this possible development.
When advisories on Subtropical Depression Seven began (noon Friday 10th October) it was located around 500 nautical miles south of the Bermuda and was expected to pass 100 nautical miles to our southeast on Sunday morning as an intensifying Tropical Storm. At this stage it was deemed a ‘Potential Threat’ and a ‘Tropical Storm Watch’ was issued. Subtropical Depression Seven developed into Subtropical Storm Fay at 6pm on October 10th around 450 nautical miles to our south. It then gained enough tropical characteristics to be deemed a pure tropical storm (TS Fay) at 9am Saturday morning.
With the storm intensifying into a strong tropical storm, and the closest point of approach getting noticeably closer to the Island for early Sunday morning on the 12th as per regular update briefings with the NHC, BWS issued a Tropical Storm Warning at 5.30am on Saturday 11th October. As a consequence of this warning, a press release was issued to by the EMO that day to advise the public to take the necessary precautions in advance of Tropical Storm Fay. Note: A large portion of the Bermuda public did not get this message as local live media is very limited on weekends; radio stations go into an automation mode. Thus the news broadcast 7pm on a Friday is critical timing.
A further update was made at 4.30pm on Saturday, when BWS in consultation with the NHC not only issued a continuation of the Tropical Storm Warning, but also a Hurricane Watch, as there was a risk that temporary hurricane force winds could affect parts of the Island (especially exposed and elevated areas) in association with heavy thunderstorms during the early morning of Sunday as Fay made its closest approach and began to accelerate away to the east-northeast. Wording in the coincident NHC Advisory #7 included the following statement:
“THE CENTER OF FAY IS EXPECTED TO PASS JUST TO THE SOUTHEAST OF BERMUDA EARLY SUNDAY MORNING. HOWEVER...ONLY A SLIGHT DEVIATION TO THE WEST OF THE FORECAST TRACK WOULD BRING THE CENTER AND THE CORE OF STRONGEST WINDS OVER BERMUDA.”
The EMO was once again advised accordingly. The Tropical Storm Warning and Hurricane Watch remained posted through Saturday night, in advance of the expected worst conditions early Sunday morning.
Wording in NHC Advisory #9 issued at 6am local time, just ahead of the worst conditions felt across the Island included: “WIND...TROPICAL STORM CONDITIONS WILL CONTINUE OVER BERMUDA THROUGH THIS MORNING. HURRICANE CONDITIONS ARE POSSIBLE OVER BERMUDA DURING THE NEXT FEW HOURS.”
Winds generally peaked at 7-8am across the Island on the morning of the 12th, a couple of hours later than initially forecast the evening before. In addition, Bermuda essentially sustained a direct hit from Fay, with its expected movement toward the northeast not occurring until after it had passed almost directly over the Bermuda area.
A chart of maximum wind speeds recorded at this time is detailed below, with the more elevated/exposed locations seeing the highest velocities: (The extreme wind conditions began to abate around 1130Z or 8.30am local time.)
Maximum sustained wind
LF Wade International Airport (MIDAS – approx. 40ft)
53.9 knots @ 1034Z
71.0 knots @ 1034Z
Causeway sensor (ultrasonic – approx. 40ft)
90.5 knots (instantaneous) @ 1103Z
RCC Bermuda Maritime Operations Centre (290ft)
84.7 knots @ 1101Z
102.2 knots @ 1101Z
St David’s AWOS (50ft)
68.9 knots @ 1050Z
100.3 knots @ 1110Z
Fort Prospect AWOS (230ft)
63.9 knots @ 1030Z
102.2 knots @ 1040Z
Commissioner’s Point, Dockyard AWOS (approx. 150ft)
75.7 knots @ 1020Z
106.7 knots @ 1020Z
Note that NHC advisory #11 finally indicated an upgrade of Tropical Storm Fay to a weak category 1 hurricane (65 knots) at 6pm on Sunday 12th October. At this point the tropical cyclone was positioned approximately 250 nautical miles to the northeast of Bermuda. Only a day or so later, at 6pm on Monday 13th, the NHC posted their final advisory on Fay as it dissipated as a post-tropical cyclone.
A wealth of media reports and photos are available on two local news websites - http://www.royalgazette.com/ and http://bernews.com/.
Considering the limited lead time available for Fay, the track and intensity forecast of this tropical storm’s onset for Bermuda was generally good, with increasing confidence as it made its final approach. Noted subtleties were:
Winds were more excessive than expected, considering Fay’s maximum sustained analysed/forecast winds of 60 knots prior, during and post Bermuda landfall. The caveat of a slight nudge to the west of the track during Fay’s final approach would essentially give Bermuda a direct hit.
Gust ratios associated with Fay were high (towards x1.6), and this is likely associated with the very active thunderstorms moving over the Island at the time.
There was evidence on the BWS Doppler Radar of tornadic activity (meso-vortices) moving across parts of the Island, especially towards the west end – see radar section later.
Damage reports: Damage from what was only an hour or so of extreme winds was widespread. There were thousands felled trees (many very mature with full foliage) as well as utility lines, and some boats were also damaged as their broke their moorings. Insurance claims are likely to have run into tens of millions of dollars (no definitive figure on this at time of writing), and some utility customers were out of power and cable. This was later compounded by further extensive outages meted out by Hurricane Gonzalo, only 6 days later.
Satellite derived wind images –Ascat satellite overpass (25km resolution, left image) at 0107Z Oct 12th clearly shows a core of strong winds on the western side of Fay approaching Bermuda:
On the right, a Windsat pass from 1102Z on the 12th shows Fay as it moved away northeast out of the Bermuda area (ringed in red):
Radar imagery - There was evidence of meso-vortices moving across Bermuda in concert with the deep convection:
Pressure - NOAA Tide Gauge at Esso Pier, St George’s:Barometer trace falling to 985.5mb/29.10inches at 0712Z (the BWS Meteograf (airport) – 985.1mb (29.09inches) at 0820UTC)
Tidal data from Esso Pier – Data suggests a maximum storm surge/tide approaching 2ft at 1124Zm around the time of mid-tide:
The Hurricane Watch was ended with the 05:30am forecast update on Sunday 12th October, with the Tropical Storm Warning remaining in place until midday, when it was superseded by a Small Craft Warning for strong winds (20kt or more) and rough seas (9ft or more).
Reclassification by the National Hurricane Center –
As is customary, the NHC specialists also compile a report on significant Tropical Cyclones. With regards to Fay, the NHC engaged in protracted communications with BWS, especially with respect to recorded wind velocities across the island, as well as the extensive data that the airport’s S-Band Doppler radar provided during the event. The main reason for this was due to the borderline hurricane status of strong Tropical Storm Fay, and whether, with extensive research of event data, the storm could be re-classified as a weak category 1 hurricane. Note that in advance of the Fay’s impact, the NHC were providing intensity forecasts via satellite derived intensity estimates (Dvorak method, Microwave imagery etc.), limited surface observations such as ocean buoys (NDBC buoy 41049, 300nm south of Bermuda), and Hurricane Hunter aircraft reconnaisance data. In effect, despite this wealth of data, the intensity forecast was a well educated estimate, but a shift of only a few knots in intensity would mean a strong 60 knots tropical storm becoming a weak 65-70 knot category 1 hurricane.
So after extensive post-analysis of all the data available, the NHC on Wednesday 3rd December were in a position to reclassify the intensity of Fay as it passed the Bermuda area. Their findings indicated that Fay was indeed already a hurricane as it brushed the southeast of the Island area during the early morning of the 12th October. It was technically a landfall with sustained winds of 70 knots. Finally, it is worth noting that Fay was the first ‘hurricane landfall’ for Bermuda since Hurricane Emily in September 1987.
Hurricane Gonzalo, 17-18th October 2014
A hurricane watch was issued by Bermuda Weather Service (BWS) at 11.30pm local time Tuesday 14th October with a hurricane warning issued at 6pm on Wednesday 15th. The BWS director, Kimberley Zuill, had been advising the local Emergency Measures Organisation (EMO) on developments since Monday 13th and along with published press releases, televised press conferences attended by the director of BWS, the Premier and the Commissioner of Police, were broadcast on Wednesday 15th and Thursday 16th. The EMO disseminated text notifications via local cell phone companies. In addition, BWS utilised social media (in advance/during/after the event) to supplement the standard forecast/warning products, and keep the local community updated. This proved to be a very effective form of communication to the public, generating an additional 2500 ‘likes’ on the BWS Facebook page https://www.facebook.com/BermudaWeatherService
The Causeway was closed (shutting off the airport and western parts of Bermuda from the rest of the island) from early morning on the 17th and the airport was closed (via NOTAM) through the 17th and into much of the 18th to facilitate the clean-up operation.
As Gonzalo began to impact Bermuda on Friday 17th, it was a still a major category 3 hurricane (100g130kt). However, as it made a close approach and subsequent landfall, it weakened slightly to a strong category 2 (95g125kt) before eventually exiting the area during the morning of Saturday 18th. The initial onset brought strengthening tropical storm force easterly winds during Friday morning. These then intensified into the hurricane force category in the afternoon and early evening as moderate to heavy rain set in. The official airfield Vaisala MIDAS wind data was lost at around 7pm as the northern eyewall and strongest easterly winds approached the area. This data was then replaced by wind speed readings from the Causeway ultrasonic sensor, located near the west end of the airfield.
The eye passage occurred around 9.30-10.15pm with winds easing variable, light to moderate across the whole Island. The persistent rain abated in what was essentially a rain-free eye. However, a rare phenomenon was observed in the overcast eye, a light ‘salt drizzle’, which was likely the vast amounts of sea spray gradually falling under gravity that had been elevated into the boundary layer by the extreme winds and seas. Winds then rapidly strengthened to hurricane force once again from the west late evening and overnight, before gradual easing into Saturday morning. In addition, further mostly moderate rainfall set in.
Remarkably, the highest gusts in the event were recorded in these westerly winds as the main southern eyewall moved across the area at around midnight local time (one would ordinarily expect the strongest winds to be on the right forward quadrant of the hurricane as Gonzalo maintained weak category 3 status). A gust of 98 knots was recorded on the Causeway (sustained at around 80 knots) and 125 knots at St David’s AWOS (sustained at 77 knots). During this wind peak, BWS had one of its storm shutters ripped off its south-facing window. So despite Gonzalo weakening from a category 3 to a category 2 during its passage across Bermuda, the strongest winds were recorded as the hurricane exited the area. The gust ratio illustrated in figure 2 varied from 1.44 within the northern eye wall to 1.62 in the southern eye wall. Suggestions as to why this occurred include the interaction of Gonzalo with a dynamic front from the west as it transited Bermuda, and the fact that it was beginning to lose its ‘pure’ tropical characteristics, slowly becoming more ‘sub-tropical’ in nature due to increased wind shear and decreasing SSTs.
Hurricane Gonzalo was a long duration event, with sustained tropical storm force winds lasting for as long as 24 hours, and hurricane force winds persisting for approximately 12 hours. The official rainfall amount at the airport for the whole event was 2.85 inches. However, this is likely to be a gross underestimate, due to much of the rain likely blowing over the gauge and not into it.
Despite a significant surge forecast to develop (an estimate of 10ft was used by NHC from a previous event – Hurricane Fabian), coastal flooding was minimal due to the eye passing (CPA) around the time of low tide. High tide occurred around 4.50pm local with an enhanced sea level of approximately 0.5ft, as per data from the NOAA Tide Gauge at Esso Pier on the north side of St George’s, see figure 12. However, with the associated storm surge beginning to impact the area at this time, the tide remained high (with in effect no low tide) and a water level differential of as much as 2.5ft was recorded at midnight before the effects of the surge on water levels began to diminish. The cumulative effects of the storm tide had less impact on the Causeway infrastructure than during Fabian, due to the direction of the of the swell/wind waves, which was more easterly rather than the forecast southeasterly. This is likely to have been due to Gonzalo pushed up against a stubborn high pressure system to the northeast, serving to back the forecast southwest winds into more of an easterly direction and giving tropical storm force winds before the forecast storm specific tropical storm force radii.
Main damage impacts across the Island, which were extensive, were further downed trees (in addition to those that came down in TC Fay, only six days prior – worse than perhaps expected due to still saturated ground from a record wet August), downed electricity and utility poles, as well as varied building damage ranging from storm shutter damage to substantial infrastructure damage (loss of roofs/collapsed walls), some of which were due to felled trees. As touched upon earlier, the Bermuda Weather Service itself was not immune from Gonzalo’s destruction, sustaining damage to the building. Most notable during the event was water ingress into a comms room and the loss of a storm shutter on the south facing side of the building. A large number of marine vessels (yachts etc.) were also grounded and damaged, even those stored on land and in ‘safe’ storm-mooring harbours. At the height of the power outage the vast majority of customers were without power (over 30,000 households). Power was only restored to all customers by the 3rd November, and some other utility (TV and Internet) customers remained without service well into November, as technicians continued to fix issues on a house-to-house basis. The Causeway sustained some damage (mostly superficial wall damage) and was re-opened during the afternoon of Saturday 18th October, along with the airport. However, the two-lane road on the Causeway was limited to one lane traffic for several days post-Gonzalo, so that restoration work on the south-facing roadside walls could be carried out. According to various media reports, the estimate of insured losses amounts to between $200 and $400 million, although this is no doubt further complicated by losses associated with Hurricane Fay just a few short days earlier.
One of the benefits of the merging of Gonzalo with a weather front over the Bermuda area was that as Gonzalo cleared northeast, the front ushered in dry and sunny weather with just moderate winds, as a ridge of high pressure gradually built in behind the front from the west. This spell of settled weather helped to expedite the restoration efforts and get Bermuda back up and running as quickly as possible.
A wealth of media reports and photos are available on two local news websites - http://www.royalgazette.com/ and http://bernews.com/; type Hurricane Gonzalo into the search boxes. These include press releases/conferences, damage pictures and clean-up efforts, including a visit by the British Royal Navy frigate HMS Argyll, which is routinely on standby in the Caribbean/West Atlantic region during hurricane season for humanitarian assistance.
Forecasting: The track and intensity forecasting of Gonzalo for Bermuda was generally very good, with high confidence provided at a long range time scale. Noted subtleties were:
Winds direction ahead of the onset of the eye on Bermuda was generally more backed (east rather than southeast) than forecast, and this may have been due to Gonzalo bumping up against the stubborn ridge to the northeast.
Another consequence of the stubborn ridge may have been the delay in Gonzalo’s onset on by a few hours – noted in the CPA in the few days running up to the event.
A side note with regard to complications in issuance of local watches and warnings – Often the synoptic pattern around Bermuda’s sub-tropical region is complex with tropical systems interacting with mid-latitude features that may induce transitioning into post-tropical near (or in rare cases over) Bermuda. In Gonzalo’s situation it would have been extremely confusing to the Bermuda public if BWS had issued a gale warning to cover the time period of gales affecting our local area due to the increased pressure gradient associated with Gonzalo pushing against the dominant Bermuda-Azores high pressure. This time frame accounted for several hours ahead of the forecast onset of tropical storm force winds associated with Gonzalo. This is a subject of much debate and concern whilst recognizing it only affects limited countries as well.
Radar imagery -
Figure left Northern eye wall radar image – coincided with strongest easterly winds. Figure right Northern eye wall clearing, temporary improvement as eye moves in.
Note that the brighter radar echoes (in red) are coincident with a spell of heavy rain reported at the airport between 5pm and 9pm local time.
Winds - Table of maximum winds (sustained and gusts) at various locations across Bermuda (ranging from approx. 40ft @ Airport/Causeway to 290 ft at Harbour Radio). Note that official reporting of sustained wind speeds in Bermuda is the 10 minute mean format. The highest gust recorded was at the St David’s AWOS, a relatively low level site: 125.4kts or 144.21mph or 230.74kph. Note that the majority of the highest wind velocities were from the WNW/NW during the southern eyewall passage, when Gonzalo was in a continued weakening phase (Cat 3 to Cat 2). The convective bands within this eyewall appeared to generate the increased gust ratio, which caused much of the most significant damage such as snapped trees, downed utility poles, roof damage and storm shutters being ripped off their mountings. Gust ratios increased from approximately 1.4 with the initial northern eyewall to as high as 1.6 with the outgoing southern eyewall. Also worth noting is the fact that there were several anemometer failures during the event, limiting a comprehensive set of forensic data:
Table of winds
Maximum sustained wind
LF Wade International Airport (MIDAS – approx. 40ft)
53.9 knots (108º) @ 2214Z 17th before anemometer failure
73.1 knots (108º) @ 2214Z 17th before anemometer failure
Causeway sensor (ultrasonic – approx. 40ft)
81 knots (WNW) @ 0250Z 18th
98 knots (WNW) @ 0250Z 18th
Commissioner’s Point, Dockyard AWOS (approx. 150ft)
94.3 knots (319º) @ 0250Z 18th
112.5 knots (323º) @ 0310Z 18th
St David’s AWOS (50ft)
78.3 knots (132º) @ 2320Z 17th
**125.4 knots (326º) @ 0310Z 18th
RCC Bermuda Maritime Operations Centre, MarOps (290ft)
109 knots in initial easterly winds before anemometer failure (max likely to be higher)
** 125.4 knots (144 mph) is the highest recorded gust in this event – stronger winds may have been recorded at the most exposed/elevated RCC MarOps site if equipment held.
Figures above: South side – Feeder rain bands & eye wall replacements; transition from Cat 3 to Cat 2 due to cooling sea surface temperatures and vertical windshear – gradual loss of tropical characteristics:
Note that these images were shared on the BWS Facebook page, in order to visually educate and update the local community on radar images that would likely be unfamiliar to them due to the relatively rare nature of a direct hit by a major hurricane. This prevented people from going outside before the bands had finished passing over, including the BDA Regiment and essential service repair teams.
Figure 11.)BWS Meteograf trace showing pressure falling to 952.3mb/28.12inches:
Pressure trace from NOAA’s Esso Pier tide station, falling to 952.3mb/28.12inches:
Tide data Lower surge impact due to timing of low tide at closest point of approach:
Onset/Cessation wind timings - Timings for onset and cessation of tropical (34kt), storm (50kt) and hurricane force (64kt) winds across the Island and Marine area (25 nautical miles offshore) based on local observations:
BWS had the rare opportunity to successfully launch a 00z weather balloon in the eye of hurricane Gonzalo.
The hurricane warning was downgraded to a Tropical Storm warning with the 0530hrs local time update on Saturday 18th October. This was then ‘Ended’ at 9am and superseded by a small craft warning which ran into Sunday morning, mainly for continued rough seas.
SST anomaly chart – Figure 17.) The NOAA/NESDIS SST anomaly chart for 20th October clearly indicated Gonzalo’s path as it approached Bermuda, with the hurricane causing some significant cooling/cold water upwelling in its wake:
Of further interest with regards to the ocean and SSTs, a local scientific institute, Bermuda Institute of Ocean Sciences (BIOS) was able to send an ocean glider into the path of Gonzalo as it approached Bermuda. Extensive data was collected including the height of underwater waves, which were recorded to be a phenomenal 150ft. For more information on BIOS’s glider programme please see the following webpages: http://www.bios.edu/news/undersea-glider-studies-hurricane-impact/
Hurricane Gonzalo was the most significant hurricane to affect Bermuda since Hurricane Fabian in 2003. Much like Fabian, the forecast for Gonzalo was mostly excellent with regards both track and intensity at a long lead time. This allowed Bermuda to comprehensively prepare for a direct hit by a major hurricane.
The forecasting, warning and preparedness was exceptional, and can be seen as a model example of how to prepare, endure and recover from a major hurricane.
However, the same cannot be said of Fay, which was not forecast especially well and had a short lead time to impact in Bermuda. The fact that Fay had such an impact on Bermuda (only six days prior) actually helped communicate the further message regarding the Gonzalo forecast/warnings and impact.
Ahead of Fay, there was an air of complacency as the Island had not suffered a direct hit by a strong tropical storm hurricane in a number of years, yet had experienced numerous tropical systems passing near enough to warrant verified watches and warnings, systems locally termed as “brushing” by the island. However, the impacts of Fay brought to the fore the fact that Bermuda can and does suffer significant tropical cyclone impacts now and again. Therefore, the community were on high alert after Fay, and paid extra attention to the watches and warnings associated with Gonzalo. In addition, it was a much stronger tropical cyclone ahead of impact, increasing to a category 4 at one stage, capturing people’s attention.
The ‘wheels’ of the forecasting/warning process ran very smoothly with Gonzalo, with plenty of lead time, allowing the community to prepare well in advance. The televised press conferences added extra weight to the warnings, with the BWS Director, Premier and Commissioner of Police communicating a strong and clear message. This was shared repeatedly with the community via various media outlets. A couple of cell phone text alerts, absent in the lead up to Fay (at least partly due to warnings going out on a weekend), were disseminated to the public helping to consolidate the message, and capture a larger swath of the community, not necessarily in tune with the more traditional forms of media such as newspapers, TV and radio. Finally, as mentioned earlier, BWS heavily used their Facebook page to provide further updates and advice to the public.
It was this combination of communication channels that undoubtedly led to excellent preparedness by the community and adherence of the strong warnings and guidance. Bermuda is very thankful that no serious injuries or loss of life were incurred as a direct result of Hurricane Gonzalo’s impact.
Bermuda’s National Conservation Officer stated that usually storm-resistant species of trees were lost much of the blame is attributed to the saturated soils from the excessive rainfall during the months beforehand. He also believes the felling of the more storm-resistant trees was the result of opposite wind directions during the two hurricane events – westerly in Fay, easterly and then westerly in Gonzalo.
One final note with regards to the rarity of two hurricane landfalls in less than a week is reiterated by the following statement which is lifted directly from the NHC’s official Tropical Cyclone Report on Hurricane Gonzalo:
“Gonzalo was the second hurricane to pass directly over Bermuda within six days; Fay made landfall on the island as a category 1 hurricane (on the Saffir-Simpson Hurricane Wind Scale) early on 12 October. This is the first time that two hurricanes made landfall on Bermuda in one hurricane season, let alone less than one week apart. Two hurricanes did affect the island within a 10-day period in 1899: during that event, a category 1 hurricane passed just northwest of Bermuda on 4 September and then a category 3 hurricane passed over or just southeast of the Island on 13 September.”