Ship Tracks Around Bermuda

2018-02-23 060220 UTC
GOES-East Geocolor imagery centred on Bermuda at 06:02 UTC, 23 February 2018. This multi-spectral product highlights low clouds (e.g. stratocumulus) as light blue shades. The brighter light blues indicate thicker low cloud.

[Watch animation]

High pressure over the western Atlantic over the last few days has been associated with persistent stratocumulus cloud cover. This typically thin layer of low-altitude cloud is a fairly common feature over the eastern half of ocean basins, and its occurrence around Bermuda is not uncommon – particularly in the cooler months.

Strong high pressure comes with subsidence. Air is compressed and warms as it descends. The descending air, with origins in the mid/upper-atmosphere is also quite dry.  This warm, dry air cannot make it all the way to the surface because it doesn’t have enough momentum to push through the cooler layer near the surface (the boundary layer). This leads to a stable layering of air that manifests as warmer drier air over cooler more humid air.

The boundary layer is often well mixed with an even distribution of heat and moisture (and can become increasingly humid over the ocean). Meanwhile, the warm air above acts like a lid over the cooler boundary layer preventing mixing between the two layers. Moisture trapped in the well mixed layer can form a sheet of stratocumulus cloud given the right conditions.

Potential temperature (the temperature that dry air would theoretically have if it was brought to the surface without external heating), Specific humidity (the mass of water vapour for every kilogram of air), and Relative Humidity (the ratio between the actual water vapour content and the water vapour content needed for saturation). Profiles measured by weather balloon at 00:00 UTC 23 February 2018 at Bermuda.

Exhaust from ships contains aerosols that act as cloud condensation nuclei. These are airborne particles that water can condense onto to form cloud droplets. The aerosols from the ships tend to form more, and smaller, cloud droplets making the clouds contaminated with ship exhaust appear brighter and thicker to satellite instruments.

The ship tracks in the animation above appear to advance from east to west or west to east, following the path of the ships. Meanwhile, another cloud enhancement appears to advance southeastward, starting from Bermuda. This track follows the wind direction in the boundary layer (unlike the ship tracks). It could be the result of aerosols from BELCO or the Incinerator acting to enhance the cloud brightness through the same mechanism as ship tracks.

High pressure and the settled weather it brings has been a theme for much of February 2018. Few mid-latitude cyclones have impacted the island so far this month. As a result, the cold air has stayed away and temperatures have remained far above normal. The lack of cyclones has also meant that precipitation totals are below normal and no days this month have experienced gale force winds. Sea surface temperatures are also far above normal as a result.

The month is expected to end with changeable weather as high pressure gives way. A cold front then takes the opportunity to push southeastward across Bermuda with some rain and showers followed by cooler air.

Further Reading:
Coakley, J. A., R. L. Bernstein, and P. A. Durkee, 1987: Effect of Ship-stack Effluents on Cloud Reflectivity. Science, 237, 1020-1022.
Hudson, J. G., 2000: Cloud Condensation Nuclei and Ship Tracks. J. Atmos. Sci., 57, 2696-2706.
Wood, R., 2012: Stratocumulus Clouds. Mon. Wea. Rev., 140, 2373-2423.


Abnormal Tides

Bermuda Esso Pier Water Levels.png
19 Sep to 12 Oct 2017 Water levels from Bermuda Esso Pier, St. George’s Island. Verified water level observations in green, preliminary water level observations in red. Forecast water level due to the predicted astronomical tide in blue. Water level is with respect to mean higher high water, where positive values indicate normally dry land is inundated. Alternative reference water level marks are on the right.

Over the weekend, the combination of the spring tide and a high amplitude ocean eddy resulted in localized coastal flooding around low-lying areas of Bermuda. Tides were running around 1.5 ft above expected levels which were already higher than normal thanks to a spring tide.

The role of the Astronomical Tides:

The astronomical tides are driven primarily by the gravitational effects of the Moon on the ocean. When the Moon is directly overhead, the water rises in response to the Moon’s gravitational pull. When the Moon is directly underfoot, the water rises again to balance the pull of the Moon on the opposite side of the Earth.

During “Spring” tides, the gravitational pull of the Sun on the oceans acts in the same direction as that from the Moon. This results in higher than normal tides and tidal ranges. Conversely, during “Neap” tides, the gravitational pull of the Sun is acting perpendicular to that of the Moon and lower than normal tides and tidal ranges can be expected. Looking at the blue line in the above figure, higher tides associated with the Spring tide can be seen around the 20th September and again last weekend, while lower tides associated with Neap tide can be seen around the 28th September.

Additionally, the Sun and Moon have to be aligned in space for their gravitational pull to act in the same direction. This manifests as a New Moon when the Moon is between the Sun and the Earth, and a Full Moon when the Earth is in between the Sun and the Moon. Both New and Full moon are associated with Spring tides. The Lunar cycle (including one Full and one New Moon) repeats roughly every 29 days and so you can expect a Spring tide a little more than every fortnight.

[More on Spring and Neap Tides]

Finally, the Moon follows an elliptical orbit around the Earth and so is closer or further away twice per orbit. Every ~7.5 Spring tides, the moon reaches its closest distance to Earth during a New or Full Moon. When the Moon is closer to Earth (perigee), the tides are slightly higher than normal. The opposite is true for when the Moon is furthest from Earth (apogee). Tides during last weekend’s Spring tide were higher than the 20th September’s Spring tide because the Moon was near/at perigee last weekend, and not during the 20th September.

The role of Ocean Eddies:

Sea surface height anomalies showing a positive anomaly greater than 30 cm near Bermuda (circled in black). This was associated with an anticyclonic eddy that intensified as it tracked southwestward toward Bermuda over the last two months.

Ever present in the ocean, eddies can manifest as regions of higher (positive) or lower (negative) sea surface height anomalies. The flow around these sea surface height anomalies is often close to balanced and so they can persist for a long time as they track across the ocean surface. These anomalies are typically small, less than 30 cm.

Typical flow around a positive sea surface height anomaly is clockwise (anticyclonic), and counter-clockwise (cyclonic) for a negative sea surface height anomaly in the northern hemisphere.

Over the weekend, a positive sea surface height anomaly associated with an anticyclonic eddy was tracking near Bermuda with amplitude estimated to be more than 30 cm (1 ft) via satellite measurements. Coinciding with the spring tide and Lunar perigee, this resulted in abnormally high water levels and some coastal inundation.

See some media mention of the tides here: Royal Gazette; Bernews (1); Bernews (2)

With sea level rise associated with climate change, it is reasonable to expect this mostly nuisance level of inundating events to occur more frequently as water level anomalies don’t have to be as extreme for flooding to occur.

Karl Passes with Minor Impact

Tropical Storm Karl failed to strengthen into a Hurricane as it passed Bermuda. Karl’s structure dramatically changed and deep convection waned in the final hours of approach. This was evidenced in aircraft reconnaissance missions that were ongoing into the storm into the early hours of Saturday. The region of peak winds shifted from the northwest quadrant to the southeast quadrant, and the center of the storm was surrounded by a gaping region of moderate winds.

Karl about 85 nm south-southwest of Bermuda with deep convection near the center beginning to organize into an eye-like feature. 1:55am local time.
Karl near closest point of approach, about 45 nm southeast of the island. deep convection redeveloping north and west of the center, but not as organized or as close to the center. 5:58am local time.

Karl passed about 45 nm southeast of the island just before 6am on Saturday with maximum sustained winds near 65mph. Winds around Bermuda peaked later in the morning as Karl moved away and winds backed to the north. Observations show sustained winds generally peaked in the 20-40 kts range with gusts up to 50 kts, particularly in elevated and exposed areas. Further, only these exposed locations saw sustained tropical storm force winds and even then, only in/around gusty showers.

Storm total rains from Karl officially reached 4.71″ at the airport. Automated personal weather stations around the island saw notably lower storm totals and this might be an artifact of measuring methods. Measuring rainfall accurately in tropical cyclones is notoriously difficult because of the accompanying high winds and I would suspect that the official total is likely closer to what actually fell.

Aside from a few delayed business openings, some transportation disruption, and isolated power outages, Bermuda fared well through Karl.

See a preliminary and unofficial list of observations around Bermuda from Wundground and BWS:


Peak Gust

Storm Total Rain

Bermuda Weather Service (Official) 41 kts (47 mph) 4.71″
MAROPS, St. George’s 50 kts (58 mph) NA
Commissioner’s Point, Sandy’s 48 kts (55 mph) NA
Pearl Island, Paget 47 kts (54 mph) NA
The Crescent 45 kts (52 mph) NA
Gilbert Hill, Smith’s 43 kts (50 mph) 1.82″
Magnolia Hall, Smith’s 41 kts (47 mph) 2.39″
Town Hill, Smith’s 41 kts (47 mph) 3.03″
Hinson’s Island, Warwick 41 kts (47 mph) 1.59″
Cardinal, Southampton 41 kts (47 mph) 2.33″
Esso Pier, St. George’s 39 kts (45 mph) NA
St. David’s, St. George’s 37 kts (42 mph) 2.53″
Chaingate Hill, Devonshire (My PWS) 34 kts (39 mph) 2.43″
Wilderness, Smith’s 34 kts (39 mph) 2.22″
McGall’s Bay, Smith’s 33 kts (38 mph) 4.15″
Devon Heights, Devonshire 30 kts (35 mph) NA
Hamilton, Pembroke 30 kts (34 mph) 2.34″
Tucker’s Town, Hamilton 27 kts (31 mph) 3.29″
Moore’s Lane, Pembroke 24 kts (28 mph) 2.28″
Ocean View, Southampton 23 kts (27 mph) 2.36″
Devonshire, Devonshire 18 kts (21 mph) 2.46″

Isolated Downpours Last Night

A stationary front, lingering in the area since the 22nd has cleared to the east and southeast this morning. Winds shifted from the southwest to the northeast through the day yesterday with the main front pushing across the island with little fan-fare. In the evening, a line of showers developed along the island extending northeast and southwest from the island. A region of convergence between northeasterly and east-northeasterly flow helped spark these showers with isolated downpours that persisted for over an hour in spots.

Distribution of rainfall from yesterday based on observations from the Bermuda Weather Service, Wunderground, and WeatherLink. This probably underestimates rain totals in Sandy’s parish as radar indicated heavier showers forming there first and being more persistent in that parish. You can really see the isolated nature of the downpours with <0.25″ across much of the eastern parishes, and over an inch in the western parishes.

Meanwhile, an area of low pressure is developing at the tail end of that lingering stationary front north of the Turks and Caicos Islands. The National Hurricane Center is monitoring this area for possible tropical cyclone formation. Their latest updates at 10am indicate a 70% chance for the formation of a tropical or subtropical cyclone in the next 5 days. However, direct impacts to Bermuda from this system appear unlikely at this stage.

Follow the Bermuda Weather Service and the National Hurricane Center for the latest official information.

Heavy Rain and Strong Storms

March 29th saw a slow moving cold front push through Bermuda from northwest to southeast. Lines of thunderstorms associated with the front tracked across the island dropping very heavy rain in a short period leading to some morning flooding issues. Lightning strikes associated with those thunderstorms resulted in isolated electricity, and communications interruptions on Tuesday. Further still, one of the thunderstorms produced a waterspout off the west end of the island that was caught on film – a fairly rare occurrence with only 5 days per year seeing either a funnel cloud or waterspout/tornado from the airport for the period 2000-2015, mainly in August or September.

Using METAR/SPECI observations and monthly climate reports from the Bermuda Weather Service this summarizes the average number of each convective phenomena by month. Annually, one day per year sees hail, 39 days see thunder, and 5 days see a funnel cloud or waterspout/tornado. Data for the period 2000-2015 inclusive.

This heavy rain event followed a string of daily record high temperatures with temperatures in the mid-70s. Highs of 75.9ºF, 76.5ºF, and 76.5ºF broke records on the 26th (previously 75.8ºF set in 1999), 27th (previously 75.2ºF set in 1999) and 29th (previously 75.0ºF set in 1989) respectively. This warmth came with unseasonably high humidity as dew point temperatures held steady near 70ºF or oppressive levels through much of that period. This warmth and moisture likely played a role in fueling the thunderstorms observed on the 29th.

Experimenting with R-code and unofficial reports from Wunderground, I’ve put together this graphical depiction of the distribution of rainfall across the island. Apparently, the heaviest rains fell in the west end where over four inches of rain was reported in spots while parts of the east end saw less than an inch. Unfortunately, there are no rain reports on Wunderground from Hamilton or Sandy’s Parishes. Gaps in report coverage limits how accurately this map represents how much rain actually fell, so I’ve indicated where each report was made.

This heavy rain event follows another event with islandwide totals of over an inch of rain just two days earlier. The combination of these events has made March 2016 unusually wet, and has officially (at the airport) brought 2016’s year to date values roughly a month’s worth of rain above average.

Distribution of rain across Bermuda shows higher storm totals to the southwest (unofficially more than four inches) and lower totals to the northeast where 0.92″ were officially measured at the Bermuda Weather Service (included in this plot.) The shaded areas are likely less reliable where there are fewer reports.

Strong Weekend Cold Front

Terra MODIS Satellite imagery for Monday morning showing cumulus field behind the cold front - typical of cold air blowing over warm water.
Terra MODIS Satellite imagery for Monday morning showing cumulus field behind the cold front streaming across the Atlantic toward Bermuda – typical of cold air blowing over warm water.

Sunday night saw the passage of a strong cold front with little fanfare save for a wind shift and increase. Winds shifted to the north and increased to strong with gale force gusts early Monday morning ushering in a much colder airmass. The airmass originated in southeastern Canada and was responsible for the first snow of the season in that region and the Northeastern United States. Temperatures fell through the morning and struggled just below 70F around midday, then continued to fall overnight and fell to record lows, which are around 65F for this time of year – 64.2F at the airport and 63.8F at my PWS.

October Cold Front 2015 Temp
October Cold Front 2015 Dew The last week’s Temperature and Dew Point trace from my PWS.

Perhaps more notable than the increased winds and lower temperatures was the drop in humidity. While midday temperatures fell from around 80F on Sunday to around 70F on Monday, the dew points fell from near 70F on Sunday to around 50F on Monday.

This much cooler and less humid weather will not stick around. Winds have since shifted to the east-northeast as high pressure extends a ridge north of Bermuda. Low pressure forming to the south of the island in the next day or two should be monitored as it will strengthen the pressure gradient across Bermuda thus increasing those east-northeasterly winds to strong again and introducing chances for rain or showers by Friday. Keep up to date with the official forecasts from the Bermuda Weather Service.

Tropical Storm Joaquin

27-30th 2015 Late September Rain Storm

Showers and rain continue in Bermuda as the stationary front continues to lift across Bermuda. Today, scattered showers and downpours are resulting from deep tropical moisture and low level convergence along a stationary front between easterly flow around high pressure to the north and southeasterly flow around newly formed Tropical Storm Joaquin, located to the southwest of the island. Storm total rainfall since the stationary front entered the picture on Saturday are 2.50″ to over 4.00″ in central and western parishes, accounting for more than half of this September’s total rainfall.

29 Sep 2015 1945UTC RGB Annotations
This evening’s RGB enhanced satellite imagery of the Western Atlantic showing Joaquin and its five-day forecast track from the National Hurricane Center – keep in mind that the 3 to 5 day period is currently an unusually low confidence forecast for Joaquin.

Joaquin has shown signs of organization today as strong northwesterly vertical wind shear abated, deep convection developed, and Joaquin continued to be embedded in a moist environment. Investigatory flights into the tropical cyclone have found that a relatively fast pace of strengthening has coincided with the improved organization. As of the 6pm advisory Joaquin had 65mph maximum 1-minute sustained winds and a minimum central pressure near 990mb.

This is a marked change from Monday when there was little forecast model support for Joaquin to strengthen. Vertical wind shear pushed the cyclone’s deep convection off to the near southeast of the surface center. This decoupled the cyclone, which means that instead of the surface circulation being vertically aligned with the mid and upper level circulations, it was tilted toward the southeast with altitude. Because low level steering flow was weak on Monday, and remains weak today, the deep convection was able to ‘pull’ the low level circulation southeastwards underneath the mid and upper level circulations while the cyclone as a whole drifted to the west. This convective re-alignment was poorly caught by the suite of forecast models yesterday.

Now that Joaquin is becoming more organized, forecast models will have a better handle on the cyclone’s short term track. A continued slow track to the west-southwest over the next 2-3 days will take Joaquin towards the northern Bahamas Islands. Joaquin’s environment will remain at least marginally favorable, if not increasingly favorable for additional development during this time. As a result, it is likely that Joaquin will become a hurricane as it approaches the Bahamas, and as of the 6pm update this evening, this scenario is now being reflected in the National Hurricane Center official forecast. While a landfall or a direct hit on any of the Bahamas Islands is not certain, tropical storm conditions are quite possible and Watches or Warnings might be required for the Northern and Central Bahamas within the next 12 hours.

The question then becomes: how far west-southwest does Joaquin go before being picked up by a quickly deepening mid-latitude system? As of the latest forecast model guidance, there appears to be several reasonable scenarios as both this mid-latitude system approaching from the northwest and the remnants of Ida approaching from the east make the steering pattern exceedingly complex – the location, amplification, and timing of both systems will have bearing on the track of Joaquin past day 3.

One scenario takes Joaquin north then northwestward into the United States East coast somewhere either in the Mid-Atlantic or Northeastern regions. Another scenario takes Joaquin northward between the US and Bermuda then out to sea. A further scenario takes Joaquin northward then northeastward toward Bermuda. Because there is so much spread in the potential track of this system, this is an unusually low confidence track forecast, particularly after day three. Hopefully, aircraft observations today will improve the accuracy of the model guidance and increase confidence in a single forecast track scenario.

Further, because Joaquin is not expected to be within 400 nm of Bermuda in the next 72 hours, the storm is not currently a threat to Bermuda at this time. However, because the potential exists for this storm to turn towards Bermuda after 72 hours it is imperative to follow updated official information from the Bermuda Weather Service and National Hurricane Center as the forecast evolves.

Japan Met Agency Radar of Typhoon Dujuan south of the Ryukyu islands over the weekend.
Japan Met Agency Radar of Typhoon Dujuan south of the Ryukyu islands over the weekend.

Recently, in the Western Pacific, Typhoon Dujuan made a direct hit on the southernmost Ryukyu islands of Japan, then a landfall in Taiwan both as a category four equivalent typhoon, followed by a second landfall in eastern China as a much weaker tropical storm. Peak wind gusts measured at 130-180mph occurred in those islands of Japan and parts of coastal northeastern Taiwan where several feet of rain also fell. Quite impressive that such measurements were possible in real time as instruments and/or communications typically fail well before winds reach that level of ferocity – a testament to the strength of local infrastructure that hopefully represents the general state of affairs in those impacted regions.