Tropical Storm Bertha (2020)

Storm Active: May 27-28

Around May 25, a weak low pressure center formed near southern Florida along a trough extending westward into the Gulf of Mexico and eastward into the Atlantic. Over the next few days, the rains associated with the system inundated portions of south Florida. More than five inches of rainfall fell in just a few hours in a few locations that day. On May 26, the low emerged over water east of northern Florida, but upper level winds were quite strong, hampering tropical development. The situation changed drastically in a short time, however, and wind shear relaxed. During the morning of May 27, thunderstorm activity blossomed near the center of circulation and the low suddenly became a tropical cyclone. Radar and buoy data indicated it already had gale force winds, so the system was classified Tropical Storm Bertha.

When it was named, Bertha was merely a couple dozen miles off the coast of South Carolina, moving northwest. It reached its peak intensity of 50 mph winds and a minimum pressure of 1004 mb at the time of landfall that same morning. Pushing inland quickly, the storm weakened to a tropical depression that afternoon. Localized flooding occurred along the storm's path over inland North Carolina into the southern Appalachians. Bertha was downgraded to a remnant low early on May 28 and dissipated later that day near western Pennsylvania.

This image shows Bertha near the time of its landfall in South Carolina.

Because it formed so close to land, Bertha spent only one day as a tropical cyclone. Nevertheless, with its formation, 2020 became the first season since 2012 to feature two named storms in the month of May.

Tropical Storm Arthur (2020)

Storm Active: May 16-19

During the second week of May, a cold front stalled over the western Atlantic, its southwestern end threading the Florida Straits. Over the next few days, wet weather prevailed in that region and a broad circulation became evident around May 14. The same day, torrential rains soaked extreme south Florida and the Keys, with lesser impacts in Cuba and the Bahamas. The system moved northeast but lacked deep convection until May 16, when the center of circulation became better defined east of the Florida peninsula. That afternoon, it was classified Tropical Depression One.

The depression became rather asymmetric that evening, with nearly all thunderstorm activity in a semicircular band east of the center. Nevertheless, winds increased enough for it to become Tropical Storm Arthur, the first named storm of the 2020 season. 2020 therefore marked the sixth consecutive year in which a storm formed before the official start of hurricane season on June 1, a new record streak. Arthur churned steadily north-northeast through May 17. Its circulation became better defined as shear lessened, but cooler ocean waters limited convective activity, leading to just a bit of strengthening that day. Rainfall began that evening in eastern North Carolina as Arthur approached.

Near the coast, the storm encountered a deep pool of warmer water in the Gulf stream and strengthened some more. On May 18, Arthur passed just east of Cape Hatteras with peak sustained winds of 50 mph. By that time, the thunderstorm activity in the western semicircle had improved significantly, leading to heavy rains over a large swath of the coast and sustained tropical storm force winds in at least the easternmost barrier islands. The mid-latitude westerlies grabbed hold of Arthur, however, and accelerated it northeastward away from land that evening. Baroclinic processes strengthened the storm a bit more as it underwent extratropical transition overnight, bringing Arthur to its peak intensity of 60 mph sustained winds and a minimum pressure of 991 mb. It turned toward the east and became post-tropical during the morning of May 19. The remnant system turned sharply south later that day and angled toward Bermuda, but weakened to an extent that it brought only intermittent showers and gusty winds as the center passed the island on May 20. It dissipated shortly afterward.

The above image shows Tropical Storm Arthur just off the coast of North Carolina.

Arthur did not quite make landfall, but it brought heavy rain to south Florida as a tropical disturbance (before formation) and the Cape Hatteras region.

Professor Quibb's Picks – 2020

My personal prediction for the 2020 North Atlantic hurricane season (written May 13, 2020) is as follows:

20 cyclones attaining tropical depression status,
18 cyclones attaining tropical storm status,
9 cyclones attaining hurricane status, and
5 cyclones attaining major hurricane status.

I predict that the 2020 Atlantic hurricane season will continue the active trend of the last few years and likely feature a well above-average number of cyclones, although uncertainty in my forecast is higher than average. Note that the average Atlantic hurricane season (1981-2010 average) has 12.1 tropical storms, 6.4 hurricanes, and 2.7 major hurricanes. The main factor in support of this claim is the potential for a neutral to negative El Niño Southern Oscillation (ENSO) index this summer and autumn. This index, a measure of equatorial sea temperature anomalies in the Pacific ocean, has an inverse correlation with Atlantic hurricane activity: negative indices (corresponding to a La Niña) tend to favor more active seasons.

Unlike the last two years, there is a great deal of spread in the model forecasts for this year's ENSO. Some models have the index remain positive, while others show a strong La Niña event developing by season's end. The dynamical models (whose average is represented by the thick red line) lean toward La Niña more than the statistical models (green), which favor using historical data for prediction over simply modeling future changes in weather. Due to a rapid plunge of Pacific equatorial sea surface temperature anomalies in the last few weeks, I'm putting more weight on the dynamical solution. This would suggest a very active season, especially since 2018 and 2019 were active despite neutral to slightly positive ENSO index conditions.

Meanwhile, Atlantic water temperatures are running uniformly warmer than average for this time of year, and this is expected to continue through the summer. No particular regions stand out though: the Gulf of Mexico, Caribbean, and subtropical Atlantic will all have moderately high anomalies, and the tropical Atlantic just a little bit less so. Expect tropical cyclone formation in many areas this year, and not quite as much emphasis on the subtropical Atlantic.

The above image shows the strength of the Saharan Air Layer (SAL) of the atmosphere as it appeared on May 12, 2020. Tropical trade winds carry dust particles and dry air from the Sahara desert over water throughout the year, but this has particular relevance to the development of tropical waves emerging off of the west African coastline: the more dry air and dust, the more thunderstorm activity in the region is suppressed. The SAL is quite strong as of now (though there is a typical decline throughout spring), so expect a late start for long-track hurricanes this year. Come September, however, other factors indicate high risk for Cape Verde hurricanes.

I'll discuss a few more smaller-scale factors in association with assigning risks to different parts of the Atlantic basin. My estimates are on a scale from 1 (least risk) to 5 (most risk):

U.S. East Coast: 4
The position of the Bermuda/Azores high pressure system correlates with the ENSO index and has a great influence on tropical cyclone tracks. I predict that this high will remain further east than typical La Niña events, increasing the risk to the east coast but lowering it for the Gulf of Mexico. The threat to the east coast will start early to the season, in July and August.

Yucatan Peninsula and Central America: 4
The western Caribbean is at higher risk for tropical cyclones than in any of the last 3 seasons, particularly when early fall rolls around. Some of the Atlantic's highest ocean temperatures will be in this region, and wind shear will be below normal. Expect some low-latitude cyclones, possibly affecting Nicaragua and Honduras.

Caribbean Islands: 4
The Caribbean, too, looks to be at greater-than-average risk this year. Once the tropical wave train gets going in earnest, look for long-track storms approaching from the east (primarily in September). The east Caribbean should be less of a hurricane graveyard than the last few seasons, so cyclones following tracks similar to Hurricane Matthew and approaching Hispaniola and Puerto Rico from the south are likelier than average.

Gulf of Mexico: 3
Waters in the Gulf are once again very warm, and some "home-grown" cyclone development is likely. Nevertheless, I forecast that the strongest cyclones will track elsewhere for most of the season. Come October, however, there is still a moderate chance for storms moving south to north from near the Yucatan Peninsula toward the United States gulf coast, similar to Hurricane Michael.

Overall, I expect the 2020 Atlantic hurricane season to feature well above-average activity. Nevertheless, this is just an amateur forecast. Individuals in hurricane-prone areas should always have emergency measures in place. For more on hurricane safety sources, see here. Remember, devastating storms can occur even in otherwise quiet seasons.

Sources: https://www.tropicaltidbits.com/analysis/models/, https://www.cpc.ncep.noaa.gov/products/analysis_monitoring/lanina/enso_evolution-status-fcsts-web.pdf, http://trackthetropics.com/saharan-air-layer-sal-tracking/

Hurricane Names List – 2020

The name list for tropical cyclones forming in the North Atlantic basin for the year 2020 is as follows:

Arthur
Bertha
Cristobal
Dolly
Edouard
Fay
Gonzalez
Hanna
Isaias
Josephine
Kyle
Laura
Marco
Nana
Omar
Paulette
Rene
Sally
Teddy
Vicky
Wilfred

This list is the same as the list for the 2014 season because no names were retired that year.

The Gravitational Assist Maneuver

On October 7, 1959, the Soviet space probe Luna 3 passed over Asia and transmitted a handful of blurry black-and-white photos back to Earth. These were humanity's first glimpse of the Moon's far side (see below). The probe was launched on a circumlunar trajectory three days prior and its initial orbit about the Earth did not take it back on a path from which it could successfully relay its valuable data. However, Luna did not propel itself onto the necessary path. This groundbreaking probe was also the first to accomplish another feat: the gravitational assist.

Since 1959, spaceflight, especially interplanetary spaceflight, has relied upon gravity assists to reach destinations all over the Solar System. Apart from Luna 3, all other examples we will consider involve using gravity assists to change orbits around the Sun, rather than some other object (such as the Earth in the case of Luna 3). The basic principle is as follows: changing orbits around the Sun requires changing spacecraft velocity relative to the Sun, known as heliocentric velocity. Conventionally, this is done using thrusters aboard the craft itself (e.g. with chemical rockets; see ion propulsion for another example). However, onboard thrusters always require ejecting mass, and heavier rockets are much more difficult and costly to launch.

Gravity assists take advantage of planets' orbital velocity and gravitational influence to alter a spacecraft's heliocentric velocity. Take the diagram below, which shows a probe flying by the planet Jupiter.

Passing close to Jupiter puts our probe on a curved trajectory about the planet. Moreover, as it falls "downhill" into the gravity well of the giant planet, it picks up speed (as indicated by the longer arrows). This gain in speed is short-lived, however, because it loses kinetic energy climbing out of the well as it departs. As a result, though the velocity vector is pointed in a different direction than before, it seems we've made no progress in increasing our probe's speed.

But this is not true: all velocities in the above diagram are relative to Jupiter, i.e. measuring the rate at which an observer on Jupiter would see the spacecraft traveling. What matters for its orbit though, is heliocentric velocity.

Suppose that, with respect to the Sun, our planet has an initial orbital velocity toward the left. Then the total heliocentric speed can increase for our spacecraft if the velocity relative to the planet is rotated to line up more closely with the planetary velocity (as indicated by the addition of arrows in the above diagram). Two of the greatest outer Solar System missions in history relied on this principle: the Voyagers.

Both Voyager 1 and 2 made use of a gravity assist at Jupiter to accelerate them to Saturn, and Voyager 2 did the same to reach Uranus and Neptune. While a slight change in direction at each planet is visible in the above diagram, the following graph better captures Voyager 2's changes in speed.

The blue curve graphs the magnitude of Voyager 2's heliocentric velocity at various distances from the Sun along its journey. Note that at each flyby of a giant planet, the probe's speed sharply increased for a short time (as it plunged into the planet's gravity well) but also was higher after each flyby than before, with the exception of after its final encounter with Neptune. The other curve shown is the Solar System escape velocity, that is, the velocity required at a specified distance from the Sun to ultimately escape its gravitational influence. Remarkably, Voyager 2 did not even have sufficient velocity to escape the Solar System until after its first flyby with Jupiter! Without gravity assists, the spacecraft would not be headed toward interstellar space today.

As indicated by Voyager 2's encounter with Neptune, gravitational assists can also reduce a spacecraft's heliocentric velocity. This is necessary for certain missions to the inner Solar System; objects launched from the Earth do not simply fall toward the Sun - they have to lose the angular momentum they inherit from our own planet!

In October 2018, the European Space Agency spacecraft BepiColombo on a mission to orbit Mercury in 2025. See here for an animation of the probe's seven-year trajectory. If the probe had been launched directly toward Mercury, the additional velocity acquired from falling into the Sun's gravity well would have made orbit impossible. Instead, the mission trajectory incorporated two Venus flybys and six Mercury flybys, all to slow down the spacecraft without the use of too much thrust.

Finally, some space missions even use encounters with planets to leave the plane of the Solar System! The Ulysses spacecraft, launched in 1990, had the goal of studying the Sun. In particular, it aimed to measure the solar wind (the flow of charged particles) and magnetic field emanating from the Sun. Unlike previous missions, Ulysses had the opportunity to study the Sun from above its poles in an orbit that was inclined 80.2° to the plane of the Solar System. A vast majority of solar system missions remain in the nearly flat plane of the Sun's equator in which the planets lie.

To achieve this unusual orbit, the probe went all the way to Jupiter just to flyby the giant planet. Jupiter's large mass allowed for a more effective gravity assist and its distance from the Sun meant that the probe was traveling slower there and the maneuver required a smaller change in velocity. Ultimately, Ulysses made a great deal of new discoveries, including that the solar magnetic field "flips" every 11 years.

Faced with the difficulties of efficient Solar System navigation, numerous space missions have utilized creative solutions involving gravity assists to reach their targets, providing another example of the spectacular innovations necessary to explore worlds beyond our own.

Sources: https://solarsystem.nasa.gov/missions/luna-03/in-depth/, https://solarsystem.nasa.gov/basics/primer/, https://www.researchgate.net/publication/228803791_Design_of_Lunar_Gravity_Assist_for_the_BepiColombo_Mission_to_Mercury, https://medium.com/teamindus/daring-gravity-assist-maneuvers-of-past-space-missions-411643cd3d55, https://solarsystem.nasa.gov/missions/ulysses/in-depth/

The Hurricane Graveyard

The Caribbean Sea appears to be a fertile area for the genesis and development of hurricanes: it lies at a tropical latitude, with warm ocean waters averaging over 80° F (26.7° C) for most of the year, including hurricane season (which extends from June to November). Warm water is essential to the development of hurricane, as it feeds warm moist air updrafts into burgeoning thunderstorms. However, during some hurricane seasons, the track map looks something like this:

The above image shows the tracks of all the tropical cyclones during the 2017 Atlantic hurricane season. Despite this season being extremely active and devastating, there is a "hole" in the map over the Caribbean, especially the eastern portion. No cyclones developed in 2017 in the eastern Caribbean and the ones that entered this region died out (though Harvey ultimately regenerated further west, with horrific consequences). This particular season was cherrypicked, but this phenomenon is sufficiently common and well-known to the meteorological community that the eastern Caribbean is sometimes known as the "Hurricane Graveyard".

Statistics confirm that the Hurricane Graveyard is more than just weather lore. Consider the diagram below, which shows the point of genesis for all known "in-season" (June to November) tropical cyclones between 1851 and 2008:

It is clear that, relative to other areas at the same latitude, the eastern Caribbean churns out far fewer tropical cyclones. What is more, tropical cyclones often unexpectedly weaken or dissipate as they pass through it, contrary to the expectations of many computer models and forecasts (however, models do fairly well at predicting the scarcity of genesis in the eastern Caribbean, see this article). So why does the Hurricane Graveyard exist?

A 2010 study examined some the factors that contribute to this phenomenon. One of these is the Caribbean low-level jet (CLLJ). Low-level jets are channels of fast-moving air in the low levels of the atmosphere near the surface (analogous to jet streams, which are much stronger and at higher altitudes). The CLLJ blows east to west over the Caribbean, with the strongest winds occurring near its center, especially during July when the CLLJ reaches its peak. The winds moderate and reach a relative minimum by October, coinciding tellingly with a somewhat greater rate of tropical cyclone formation in the east Caribbean.

The above image shows the low-level winds in the Caribbean region averaged over the month July 2006. As the diagram indicates, wind speeds in excess of 10 m/s (22 mph) are typical during that time of year. These winds are due in part to the Azores-Bermuda high pressure system, a persistent area of dry, sinking air over the central subtropical Atlantic around which air flows clockwise. The southwestern portion of clockwise flow is evident in the diagram above. But why do horizontal winds of this sort inhibit tropical cyclone development? Further, the peak winds occur toward the center of the Caribbean, whereas our area of interest lies to the east, so what gives?

The key concept explaining the hurricane graveyard is low-level divergence. Divergence is simply a measure of to what degree air (or any fluid) is flowing out of a region. Two schematic vertical cross-sections of the atmosphere are shown below.

In the first, there is low-level divergence, as air is flowing outward from the central region. Since any outgoing air must be replaced, this air can only come from above, creating downdrafts and convergence (the opposite) at the atmosphere's upper levels. The reverse occurs in the second diagram. One may note that our diagram of Caribbean winds does not show air moving outward from any central point. However, the CLLJ accelerates moving east to west across the eastern portion of the sea, so it is still a situation where more air is leaving this region in the lower levels than is entering it. Hence the eastern Caribbean exhibits low-level divergence!

Crucially, low-level divergence is the opposite of what hurricanes need to thrive. Thunderstorm activity is driven by rising moist air and killed by sinking dry air. The divergence caused by the CCLJ often causes the thunderstorms associated with tropical cyclones or disturbances to collapse when they pass over the region, at last accounting for the Hurricane Graveyard.

Recent analyses have shed even more light on this phenomenon. It peaks in July when the CCLJ does, diminishing toward the end of hurricane season. Indeed, the rate of cyclonogenesis in the eastern Caribbean more closely resembles that of the rest of the basin from October onward. In addition, the Hurricane Graveyard effect is more pronounced in summers with an El Niño, when the Azores-Bermuda high is farther west and stronger, leading to a more powerful CCLJ. This effect illustrates how there is a great deal more to hurricane formation than humid air and warm water. Several other atmospheric factors, including low-level divergence, feature crucially in understanding and forecasting the development of tropical cyclones.

Sources: https://journals.ametsoc.org/doi/abs/10.1175/2009BAMS2822.1, https://journals.ametsoc.org/doi/full/10.1175/WAF-D-13-00008.1, https://journals.ametsoc.org/doi/pdf/10.1175/2011JCLI4176.1, http://www.atmo.arizona.edu/students/courselinks/fall16/atmo336/lectures/sec1/winds_fall15.html

Atmospheric Rivers

The transport of water vapor in the atmosphere is responsible for the world's weather. As such, tracking the movement of water-laden air masses is essential to weather forecasting. By the mid-1990's, satellite measurements had documented that the flow of moisture through the atmosphere often occurs in very high volumes through very narrow channels. These phenomena were named atmospheric rivers, because the amount of water they transport is comparable to that of Earth's largest rivers.

The above graphic (click to enlarge) summarizes the basics of atmospheric rivers. The most well-known example, as indicated above, delivers water vapor from near the equator to the western coast of north America, especially California. This phenomenon is the source of a large portion, if not a majority, of precipitation in these regions when moist air is forced upward over the Sierra Nevada mountains.

The formation of a specific atmospheric river event often occurs when a strong extratropical cyclone moves away from the equator toward higher latitudes, dragging warm equatorial air in its wake. This paper analyses the role such cyclones have in driving atmospheric rivers, focusing on a case study of a powerful February 2002 cyclone. This cyclone formed in the north Atlantic and moved northeastward, passing west of the British Isles before ultimately dissipating in the Barents Sea north of Scandinavia. At its peak intensity, it bottomed out at 935 mb, a pressure comparable to that of a category 4 hurricane!

The above shows the cloud pattern of the cyclone on an infrared satellite view (top) and a diagram showing total column water vapor (TCWV, bottom) in the same region. Unlike for hurricanes, most of the extratropical storm's water vapor is concentrated in a linear feature far from the center and extending south and west. The counterclockwise rotation of the large circulation in the northern hemisphere draws a long train of warm, tropical air northward, and it is here that the atmospheric river sets up.

In practice, meteorologists detect atmospheric rivers using satellite imagery, which can detect the transport of water vapor in the atmosphere, and the threshold for being an atmospheric river depends on the integrated water vapor transport (IVT) value in a given region. This is calculated by adding up the flow of moisture in different layers of the atmosphere. Using these tools, the frequency/severity of such events across the world has recently been analyzed.

The above map shows the percentage of severe surface precipitation events are associated with atmospheric rivers in different locations across the world over the period 1997-2014. Only precipitation events ranking in the top 2% for a given region are considered. Therefore, the map shows where atmospheric rivers have the greatest impact: mostly over the ocean, but also in portions of western North America, southern South America, central Asia, and eastern Australia. They also are associated with extreme wind events over many of the same regions. As a result, atmospheric rivers are responsible for many of the most damaging flooding events across the globe. Better understanding their formation and evolution will improve our ability to forecast and respond to flooding events in the future.

Sources: https://web.archive.org/web/20100610035058/http://paos.colorado.edu/~dcn/ATOC6020/papers/AtmosphericRivers_94GL01710.pdf, https://noaa.gov, https://journals.ametsoc.org/doi/pdf/10.1175/BAMS-D-14-00031.1, http://cw3e.ucsd.edu/wp-content/uploads/2018/02/WaliserGuan-ngeo2894.pdf

Upper Atmospheric Clouds

Weather is caused by the atmosphere. However, most of what we know as weather is limited to the atmosphere's lowest layer, the troposphere. The height of the troposphere varies with latitude on the Earth, from under 4 miles (6 km) in the coldest polar regions to over 10 miles (16 km) at the equator. The distinguishing feature of this layer that allows weather to occur is cloud formation.

Cloud formation occurs when water, evaporated from the ground into gaseous water vapor, condenses into small water droplets in the atmosphere. However, the water vapor-laden air must have some means by which it is transported upward. One such means is orthographic cloud formation.

This occurs when the wind direction pushes air up the slope of a topographical feature such as a mountain. As air rises, it expands and cools. Since cooler air holds less water vapor, condensation begins, clouds form, and precipitation can follow. Another impact of this process is that the opposite side of the mountain from the prevailing wind can be very dry.

However, clouds of course also form in the absence of mountainous terrain. This occurs when a parcel of air near the surface, containing evaporated water, is heated by the warm sunlit ground. Warmer than the adjacent atmosphere, it rises. As the air rises, its pressure drops in accord with that of its surroundings (air pressure decreases with altitude, a phenomenon mountain climbers are familiar with). By the ideal gas law, the pressure of a gas is proportional to its temperature. Therefore, a rising parcel of air cools.

The tallest clouds form in the case of absolute instability, illustrated above (click to enlarge). The parcel of air begins at the ground at a temperature of 40° C. As it rises, it cools. However, in the unstable case, the air temperature of the ambient atmosphere drops more quickly than the temperature in the rising parcel of air (the rate of its cooling is known as the dry adiabatic lapse rate). As a result, the cooling air nevertheless remains warmer than its surroundings and continues to rise. Eventually, its becomes too cold to hold all of the water vapor and condensation begins at the condensation level. Condensation releases heat to the atmosphere, so the temperature decrease of the parcel slows (to what is known as the wet adiabatic lapse rate). The resulting condensation forms a cloud, whose lower extremity is the condensation level.

At the top of the above diagram, the parcel temperature is still much higher than the atmospheric temperature, so the air rises yet more, continuing to increase the height of the cloud on the way. In situations such as this, a cumulonimbus cloud may form.

Cumulonimbus can extend from a few thousand feet from the ground over 10 miles (16 km) upward. It is these clouds that are responsible for many types of extreme weather, including severe thunderstorms, hail, and even tornadoes. However, even for these enormous clouds, there is an limit to their upward growth: the tropopause. The tropopause is the boundary between the troposphere below and the next layer of the atmosphere, the stratosphere, above. As indicated at the beginning of the post, the height of the tropopause varies with latitude. It is characterized by a reversal of the decline in temperature with height that takes place up to this point. Beginning at the tropopause and into the stratosphere, temperature again begins to increase with height (as shown near the bottom of the figure below).

Unlike in the troposphere, where the only significant source of heat was the sun-heated surface below, the stratosphere derives heat from another source: the absorption of solar ultraviolet radiation by the ozone layer. Oxygen and ozone (O₃) molecules interact with incoming ultraviolet rays, impeding their passage to the Earth's surface and therefore protecting it from most of this harmful radiation. In the process, they absorb heat. The result of all this is that air just above the tropopause is always stable: temperature increases with height. Rising parcels of air do not penetrate much past this point. Rather, they spread out laterally, forming the anvil top characteristic of cumulonimbus clouds and illustrated above. By virtue of their momentum, a few parcels enter a bit into the stratosphere; this is the "overshooting top" in the diagram.

However, in rare cases, clouds can form outside the troposphere. For example, polar stratospheric clouds (PSCs) may sometimes form in high latitude regions. They occur in the lower stratosphere, at altitudes varying from 6 to 15 miles (10-25 km), where temperatures plummet to below -100° F. At these especially low temperatures, water vapor can condense directly into ice crystals. The more spectacular PSCs are the nacreous clouds, which are composed entirely of these crystals. They are most visible just after sunset when the Sun has disappeared below the horizon on the ground but still illuminates the stratosphere. The ice crystals diffract light, producing beautiful iridescent displays.

This image shows nacreous clouds over McMurdo Station, Antarctica. Though they are most often seen over that continent, they also appear occasionally over northern Europe, North America, and Russia. PSCs are of scientific interest because they can serve as sites for chemical reactions that destroy ozone. Understanding the climatology of these unusual clouds helps to monitor the regional variation of ozone in polar regions.

In addition to PSCs, there is another type of upper atmospheric cloud that occurs even at even higher altitudes. Noctilucent clouds, so named because they are only visible at night well after the Sun has passed below the horizon, form in the mesosphere. This layer is defined as being the second region where temperature decreases with height. The clouds form generally around 50 miles (80 km) above the surface. Though not fully understood, it is thought the formation of noctilucent clouds is due to meteoric dust. This dust is the result of small objects from space breaking up upon entering Earth's atmosphere and serves as a site for the formation of ice crystals. Though the amount of moisture water vapor this high is minuscule, the temperature here can plummet to a frigid -180° F at high latitudes, supporting ice crystal formation.

The above image, taken from Nunivak Island, Alaska, shows noctilucent clouds after sunset. These clouds are at such a high altitude that they reflect sunlight from below the horizon back down to Earth. Studying the locations where these clouds form can reveal concentrations of dust (meteoric or from other sources such as volcanoes) in the mesosphere as well as shed light on how the upper atmosphere changes with the climate. Apart from their scientific value, upper atmospheric clouds join the aurorae as some of the most beautiful displays to be found in polar skies.

Sources: The Encyclopedia of Weather and Climate Change by Juliane Fry, https://www.wunderground.com/cat6/Methane-Giving-Noctilucent-Clouds-Boost?cm_ven=hp-slot-1, http://www.richhoffmanclass.com/chapter4.html, https://www.nasa.gov/mission_pages/sunearth/science/atmosphere-layers2.html, https://www.atoptics.co.uk, https://www.nasa.gov/multimedia/imagegallery/image_feature_680.html, https://www.sciencedaily.com/releases/2014/04/140411091939.htm

2019 Season Summary

The 2019 Atlantic hurricane season had above-average activity, with a total of

20 cyclones attaining tropical depression status,
18 cyclones attaining tropical storm status,
6 cyclones attaining hurricane status, and
3 cyclones attaining major hurricane status.

Before the beginning of the season, I predicted that there would be

15 cyclones attaining tropical depression status,
14 cyclones attaining tropical storm status,
6 cyclones attaining hurricane status, and
3 cyclones attaining major hurricane status.



The average number of named storms, hurricanes, and major hurricanes for an Atlantic hurricane season (over the 30-year period 1981-2010) are 12.1, 6.4, and 2.7, respectively. The 2019 season was much above average in the number of tropical storms (in fact, it was only the sixth year since the advent of the modern naming system that the "S" name was used), but near average in the hurricane and major hurricane categories. This stemmed in part from the fact that the season featured many short-lived, weak tropical storms and a few powerful hurricanes. The ACE value for the 2019 season (which accounts for duration and intensity of storms as well as number) was around 130, just above the average. Apart from the addition of a few extra short-lived cyclones, my predictions were on target.

The El Niño event initially forecast to continue (see the first image in this post) through the 2019 summer/fall died out by July, transitioning to an ENSO neutral/slightly positive state that continued through November. Since an El Niño event typically suppresses hurricane activity, this factor alone would suggest an average to slightly below average season. Indeed, 2019 did display some of the hallmarks of an El Niño season, including storm tracks curving away from the north American mainland and wind shear in the main development region (the Caribbean and tropical western Atlantic). However, the "tropical wave train" off of Africa was quite vigorous for most of the season, and even extended later than usual into October (with Tropical Depression Fifteen among the latest forming tropical waves off of Africa ever recorded). Also offsetting the residual El Niño was the continuing ocean warmth of the subtropical Atlantic, which contributed to prolific development; nearly half of the season's storms formed in the subtropical Atlantic.

The most damaging storms of the 2019 season were Hurricane Dorian and Tropical Storm Imelda. Dorian devastated the Bahamas, especially Abacos Island and Grand Bahama, as a category 5 hurricane. It was the strongest landfalling storm on record for the region and was tied for the most intense landfalling hurricane by windspeed ever recorded in the Atlantic (with the Labor Day hurricane of 1935). Imelda was only a tropical storm, but its slow motion over southeast Texas caused a major flooding event similar to, though slightly less severe than, those associated with Harvey and Florence in the previous two years. Some other notable facts and records from the 2019 season include:

• Seven of the eighteen named storms of the 2019 season lasted less than 24 hours, a new record
• Subtropical Storm Andrea formed on May 20, marking the fifth consecutive year in which a named storm developed before the official start of the season on June 1 (also a new record)
• Dorian was the strongest Atlantic hurricane ever recorded so far north at its latitude of peak intensity, 26.6° N
• When Lorenzo became a category 5 hurricane, it marked only the seventh time on record that multiple category 5 hurricanes were recorded in a season; Lorenzo was the easternmost forming category 5 on record
• Pablo strengthened into a hurricane farther north and east than any cyclone previously observed, at 42.8° N, 18.3° W, only a few hundred miles west of the coast of Spain

Overall, the 2019 season was somewhat above average (though near average in number of hurricanes and major hurricanes), with Hurricane Dorian the strongest and most damaging.

Tropical Storm Sebastien (2019)

Storm Active: November 19-24

Around November 15, a trough of low pressure formed over the central Atlantic, generating a diffuse area of shower activity well to the east-northeast of the Lesser Antilles. The system moved steadily northwestward over the next few days and developed a weak low pressure center on November 17. Convective activity concentrated some near the center soon after, though upper-level winds and dry air out of the west confined most thunderstorm activity to the disturbance’s eastern side. During the morning of November 19, satellite data indicated a closed circulation was present and the low was upgraded to Tropical Storm Sebastien.

The next day, the cyclone turned northward as it rounded the edge of a subtropical ridge and began to feel the influence of an approaching front from the west. Sebastien faced heavy shear but still managed to strengthen some due to favorable upper-level divergence. That day, it continued its turn to the northeast, accelerating some as it did so. Convection was rather disorganized, but the system maintained its identity as it moved ahead of the advancing frontal boundary.

By November 22, Sebastien was beginning to exhibit some extratropical characteristics, but its inner core was still clearly that of a tropical system as it sped up even further. Under the cyclone, ocean heat content rapidly diminished, but it did not weaken, and in fact reached its peak intensity of 65 mph winds and a pressure of 994 mb the next day. Sebastien was already bringing heavy surf and high winds to the Azores by this time, and it rocketed across the westernmost islands during the evening of November 24. At last, its quick forward speed, coupled with cold ocean waters and frontal interaction, caused the storm to fully transition to extratropical later that night. The remnants of Sebastien merged with another system near Ireland a little over a day later.

The above image shows Tropical Storm Sebastien on November 24 shortly before extratropical transition.

The track and evolution of Sebastien were not well anticipated by computer models. While the general recurvature was agreed upon, initial model runs did not indicate Sebastien would remain tropical for as long as it did, nor that it would be able to maintain strong tropical storm intensity.