A massive dust storm that formed over the Sahara Desert early this week has now pushed out over the tropical Atlantic, and will sharply reduce the odds of tropical storm formation during the first week of August. The dust is accompanied by a large amount of dry air, which is making the Saharan Air Layer (SAL) much drier than usual this week. June and July are the peak months for dust storms in the Southwest Sahara, and this week’s dust storm is a typical one for this time of year. Due in large part to all the dry and dusty air predicted to dominate the tropical Atlantic over the next seven days, none of the reliable computer models is predicting Atlantic tropical cyclone formation during the first week of August.
Figure 1. A massive dust storm moves off the coast of Africa in this MODIS image taken at 1:40 UTC July 30, 2013. Image credit: NASA.
How dust affects hurricanes
Saharan dust can affect hurricane activity in several ways:
1) Dust acts as a shield which keeps sunlight from reaching the surface. Thus, large amounts of dust can keep the sea surface temperatures up to 1°C cooler than average in the hurricane Main Development Region (MDR) from the coast of Africa to the Caribbean, providing hurricanes with less energy to form and grow. Ocean temperatures in the MDR are currently 0.7°F above average, and this anomaly should cool this week as the dust blocks sunlight.
2) The Saharan Air Layer (SAL) is a layer of dry, dusty Saharan air that rides up over the low-level moist air over the tropical Atlantic. At the boundary between the SAL and low-level moist air where the trade winds blow is the trade wind inversion–a region of the atmosphere where the temperature increases with height. Since atmospheric temperature normally decreases with height, this “inversion” acts to but the brakes on any thunderstorms that try to punch through it. This happens because the air in a thunderstorm’s updraft suddenly encounters a region where the updraft air is cooler and less buoyant than the surrounding air, and thus will not be able to keep moving upward. The dust in the SAL absorbs solar radiation, which heats the air in the trade wind inversion. This makes the inversion stronger, which inhibits the thunderstorms that power a hurricane.
3) Dust may also act to produce more clouds, but this effect needs much more study. If the dust particles are of the right size to serve as “condensation nuclei”–centers around which raindrops can form and grow–the dust can act to make more clouds. Thus, dust could potentially aid in the formation and intensification of hurricanes. However, if the dust acts to make more low-level clouds over the tropical Atlantic, this will reduce the amount of sunlight reaching the ocean, cooling the sea surface temperatures and discouraging hurricane formation (Kaufman et al., 2005.)
Dust in Africa’s Sahel region and Atlantic hurricane activity
The summertime dust that affects Atlantic tropical storms originates over the southwestern Sahara (18° – 22° N) and the northwestern Sahel (15° – 18° N) (Figure 3.) The dust from the Southwest Sahara stays relatively constant from year to year, but the dust from the Northwest Sahel varies significantly, so understanding this variation may be a key factor in improving our forecasts of seasonal hurricane activity in the Atlantic. The amount of dust that gets transported over the Atlantic depends on a mix of three main factors: the large scale and local scale weather patterns (windy weather transports more dust), how wet the current rainy season is (wet weather will wash out dust before it gets transported over the Atlantic), and how dry and drought-damaged the soil is. The level of drought experienced in the northwestern Sahel during the previous year is the key factor of the three in determining how much dust gets transported over the Atlantic during hurricane season, according to a January 2004 study published in Geophysical Research Letters published by C. Moulin and I. Chiapello. In 2012 (Figure 3), precipitation across the northwestern Sahel was much above average, which should result in less dust than usual over the Atlantic this fall, increasing the odds of a busy 2013 hurricane season.
Figure 3. Rainfall over the Northwest Sahel region of Africa was about 200% of average during the 2012 rainy season. The heavy rains promoted vigorous vegetation growth in 2013, resulting in less bare ground capable of generating dust. Image credit: NOAA/Climate Prediction Center.
Dr. Amato Evan published a study in Science magazine March 2009 showing that 69% of the increase in Atlantic sea surface temperatures over the past 26 years could be attributed to decreases in the amount of dust in the atmosphere.
Kaufman, Y. J., I. Koren, L. A. Remer, D. Rosenfeld, and Y. Rudich, 2005a: The effect of smoke, dust, and pollution aerosol on shallow cloud development over the Atlantic Ocean. Proc. Natl. Acad. Sci. USA, 102, 11 207–11 212.
In the Central Pacific, Tropical Storm Flossie, a strong tropical storm with 65 mph winds, is headed west at 20 mph towards Hawaii. Satellite images show that Flossie is maintaining a modest area of heavy thunderstorms that are well-organized. The storm is over waters of 25°C, which is about 1°C below the water temperature typically needed to sustain a tropical storm. Flossie peaked in intensity Saturday morning, when the storm had 70 mph winds. As Flossie approaches the Big Island of Hawaii on Monday, these waters will warm to 26°C, but wind shear is expected to be in the moderate range, which should keep Flossie from strengthening. Dry air aloft will likely cause some weakening before landfall Monday morning, and Flossie will likely have top winds of 45 – 55 mph when is passes through the Hawaiian Islands. Flossie’s main threat will be heavy rains, with 6 – 10″ expected over The Big Island and Maui County, and 4 – 8″ in Oahu. Rains of this magnitude are capable of causing dangerous flash flooding and mudslides. Sunday’s 11 am EDT wind probability forecast from the Central Pacific Hurricane Center gave Hilo on the Big Island a 33% chance of experiencing sustained tropical storm force winds of 39 mph or greater from Flossie. These odds were 32% for Honolulu and 41% for Kahului.
Figure 1. MODIS satellite image of Tropical Storm Flossie taken at approximately 5 pm EDT Saturday July 27, 2013. At the time, Flossie had top winds near 50 mph. Image credit: NASA.
Tropical storms are uncommon in Hawaii
On average, between four and five tropical cyclones are observed in the Central Pacific every year. This number has ranged from zero, most recently as 1979, to as many as eleven in 1992 and 1994. August is the peak month, followed by July, then September. Tropical storms and hurricanes are uncommon in the Hawaiian Islands. Only eight named storms have impacted Hawaii in the 34 year period 1979–2012, an average of one storm every four years. Since 1949, the Hawaiian Islands received a direct hit from just two hurricanes–Dot in 1959, and Iniki in 1992. Both hit the island of Kauai. Only one tropical storm has hit the islands since 1949–an unnamed 1958 storm that hit the Big Island. A brief summary of the three most significant hurricanes to affect Hawaii in modern times:
September 1992: Hurricane Iniki was the strongest, deadliest, and most damaging hurricane to affect Hawaii since records began. It hit the island of Kauai as a Category 4 on September 11, killing six and causing $2 billion in damage.
November 1982: Hurricane Iwa was one of Hawaii’s most damaging hurricanes. Although it was only a Category 1 storm, it passed just miles west of Kauai, moving at a speed of nearly 50 miles per hour (80 km/h). Iwa killed one person and did $250 million in damage, making it the second most damaging hurricane to ever hit Hawaii. All the islands reported some surf damage along their southwest facing shores, and wind damage was widespread on Kauai.
August 1959: Hurricane Dot entered the Central Pacific as a Category 4 hurricane just south of Hawaii, but weakened to a Category 1 storm before making landfall on Kauai. Dot brought sustained winds of 81 mph with gusts to 103 mph to Kilauea Light. Damage was in excess of $6 million. No Dot-related deaths were recorded.
Figure 2. Tracks of all tropical storms and hurricanes to pass within 100 miles of the Hawaiian Islands, 1949 – 2012. During that time span, the Hawaiian Islands received a direct hit from just two hurricanes–Dot in 1959, and Iniki in 1992. Both hit the island of Kauai. One tropical storm also hit, and unnamed 1958 storm that hit the Big Island of Hawaii. Image credit: NOAA/CSC.
Remains of Dorian are worth watching
The remains of Tropical Storm Dorian will be passing just north of northern Lesser Antilles Islands today and just north of Puerto Rico tonight. Satellite images show no signs of a surface circulation, and just a moderate area of heavy thunderstorms associated with the storm. AIr Force hurricane hunter aircraft are on call to investigate Dorian’s remains on Sunday afternoon and Monday afternoon, if necessary. In their 8 am EDT Tropical Weather Outlook, NHC gave Dorian’s remains a 20% chance of regenerating by Tuesday.
Extreme heat wave in Europe
An extreme heat wave is baking Europe today, and at least five countries have a chance at setting a new all-time national heat record. The most likely candidate is Liechtenstein, where the forecast for Balzers Sunday calls for a high of 95°F. According to wunderground’s International Records database maintained by our weather historian, Christopher C. Burt, the current all-time heat record for Liechtenstein is 36°C (96.8°F) set at Vaduz on August 13, 2003.
The season’s fourth named storm, Tropical Storm Dorian, is here. Born from a strong tropical wave that moved off the coast of Africa on Monday, Dorian formed unusually far east for so early in the season, at longitude 29.9°W. Only Hurricane Bertha of 2008, which became a tropical storm at 22.9°W longitude on July 3, formed farther to the east so early in the year. Satellite images show that Dorian is a small but well-organized system with a moderate amount of heavy thunderstorms. A large area of dry air lies to Dorian’s west, as seen on water vapor satellite images, but Dorian has moistened its environment enough that this dry air should not interfere with development for the next day. Dorian is under a low 5 – 10 knots of wind shear, which will tend to allow slow development. Ocean temperatures are barely adequate for maintaining strength of a tropical storm, about 26.5°C.
Figure 1. MODIS satellite image of Tropical Storm Dorian taken at approximately 8 am EDT July 24, 2013. At the time, Dorian had top winds near 50 mph. Image credit: NASA.
Forecast for Dorian
The SHIPS model predicts that wind shear will stay in the low range through Thursday, then rise to the moderate range Friday through Monday. Ocean temperatures will fall to 25 – 26°C Wednesday night through Thursday night, which may induce some weakening of Dorian. Thereafter, ocean temperatures will rise again, but wind shear will rise. This increase in wind shear will be capable of causing weakening, since there will still be a large area of dry air to Dorian’s west that the shear may be able to bring into Dorian’s core. Given its small size, Dorian is capable of relatively large changes in intensity in a short amount of time, and it would not surprise me if the storm dissipated by the end of the week–or became a Category 1 hurricane. However, the official NHC forecast of a tropical storm passing just north of the Lesser Antilles on Sunday is the most likely outcome; the 11 am wind probability forecast from NHC gave Dorian a 6% chance of being a hurricane at that time. Dorian should maintain a west-northwest track through the week, and spread heavy rains and gusty winds to the northern Lesser Antilles Islands beginning on Sunday. The usually reliable European model (ECMWF) has Dorian passing several hundred miles to the north of the Lesser Antilles Islands, while the other models show Dorian passing closer, within 100 miles. It currently appears that Dorian will be a potential threat to the Bahama Islands, Bermuda, and the U.S. East Coast next week. There will be a trough of low pressure capable of recurving Dorian out to sea before the storm reaches the Bahamas and U.S., but this trough is currently depicted as being fairly weak, reducing the chances of Dorian missing the Bahamas and U.S. East Coast.
Figure 2. Tracks of all Atlantic tropical depressions, tropical storms, and hurricanes (tropical cyclones) occurring in the months of June and July off the coast of Africa. Only Bertha of 2008 became a named storm farther east so early in the year, compared to Tropical Storm Dorian. Reliable satellite records of Eastern Atlantic tropical cyclones go back to 1966. Image credit: NOAA/CSC.
June 2013 was the 15th warmest June in the contiguous U.S. since record keeping began in 1895, said NOAA’s National Climatic Data Center (NCDC) in their latest State of the Climate report. Six Southwest U.S. states had a top-ten warmest June on record, and no states recorded a significantly below-average June for temperatures. Over three times as many record warm highs and lows occurred than record cold highs and lows during June. For the year-to-date period January – June, both temperature and precipitation over the contiguous U.S. have been above normal, ranking in the upper 33% and 23% of years, respectively.
According to NOAA’s U.S. Climate Extremes Index (CEI), which tracks the percentage area of the contiguous U.S. experiencing top-10% and bottom-10% extremes in temperature, precipitation, and drought, June extremes were about 10% below average, and the year-to-date period January – June 2013 has been 20% below average.
Figure 1. Historical temperature ranking for the U.S. for June 2013. Six Southwest U.S. states had a top-ten warmest June on record, and no states recorded a significantly below-average June for temperatures. Image credit: National Climatic Data Center (NCDC).
Wet June on the East Coast raises hurricane flood risk
It was a very June for the contiguous U.S., ranking as the 13th wettest June since 1895. New Jersey and Delaware had their wettest June on record, and sixteen other eastern states had a top-ten wettest June. The very wet June has brought some of the highest soil moisture levels ever recorded for July along much of the coast from Florida to Maine, increasing the chances of extreme flooding should this region receive a hit from a tropical storm or hurricane during the coming peak months of hurricane season. The latest 2-week forecast from the GFS model keeps the East Coast under a wetter-than-average weather pattern into early August, and the latest 1-month and 3-month precipitation outlooks from NOAA’s Climate Prediction Center also give above-average chances of wetter than average conditions. Lake Okeechobee in Florida is 1.4′ above average for this time of year, and 5′ higher than two years ago. While this still puts the lake 1.2′ below what is considered high water, Lake Okeechobee water levels will need to be watched as we head into the peak part of hurricane season.
Figure 2. Historical precipitation ranking for the U.S. for June 2013. New Jersey and Delaware had their wettest June on record, and sixteen other eastern states had a top-ten wettest June on record. Utah had its driest June on record, and Arizona, Colorado, and Wyoming had a top-ten driest June. Image credit: National Climatic Data Center (NCDC).
Figure 3. Soil moisture for July 14, 2013, expressed as percent average of the soil moisture observed between 1916 – 2004. Portions of Florida, Georgia, South Carolina, North Carolina, Virginia, Maryland, New York, Vermont, and New Hampshire are near their highest soil moisture levels on record for this time of year, increasing the odds of extreme flooding in those states should a tropical storm or hurricane hit this year. Image credit: University of Washington Variable Infiltration Capacity Macro-scale Hydrological Model, which includes soil moisture, snow water equivalent, and runoff.
Drought conditions remained relatively unchanged during June. According to the July 9 Drought Monitor report, about 45% of the contiguous U.S. is still in moderate or greater drought, compared to 44% at the beginning of June. The U.S. Seasonal Drought Outlook issued on June 21 calls for little overall change in the U.S. area covered by drought conditions during the remainder of summer. Approximately 1.2 million acres of land burned in the U.S. during June, which is above average. However, the year-to-date total acreage burned is the second lowest in the past ten years.
Quiet in the Atlantic
There are no tropical cyclone threat areas in the Atlantic to discuss today, and none of the reliable models for tropical cyclone formation is predicting development during the coming seven days.
Earth’s deadliest natural disaster so far in 2013 is the deadly flooding in India’s Himalayan Uttarakhand region, where torrential monsoon rains have killed at least 556 people, with hundreds more feared dead. At least 5,000 people are missing. According to the Indian Meteorological Department, Uttarakhand received more than three times (329%) of its normal June rainfall from June 1 – 21, and rainfall was 847% of normal during the week June 13 – 19. Satellite estimates indicate that more than 20″ (508 mm) or rain fell in a 7-day period from June 11 – 17 over some regions of Uttarakhand, which lies just to the west of Nepal in the Himalayas. According to Dr. Dave Petley’s Landslide Blog, Earth’s deadliest landslide since the August 2010 Zhouqu landslide in China hit Uttarakhand’s Hindu shrine in Kedarnath, which is just a short distance from the snout of two mountain glaciers. The shrine is an important pilgrimage destination this time of year, and was packed with visitors. Hindu devotees visit Uttarakhand in huge numbers for the char-dham yatra, or a pilgrimage to the four holy sites of Gangotri, Kedarnath, Yamnotri and Badrinath. Apparently, heavy rainfall triggered a collapse event on the mountain above Kedarnath, which turned into a debris flow downstream that struck the town. The main temple was heavily damaged, and numerous building in the town were demolished.
Figure 1. Indo-Tibetan Border Police (ITBP) arrive to rescue stranded Sikh devotees from Hemkunt Sahib Gurudwara, a religious Sikh temple, to a safe place in Chamoli district, in northern Indian state of Uttarakhand, India, Monday, June 17, 2013. AP photo.
Figure 2. Satellite-estimated rainfall for the 7-day period June 11 – 17, 2013, from NASA’s TRMM satellite exceeded 20 inches (508 mm) over portions of India’s Uttarakhand province, leading to catastrophic floods. Image credit: NASA.
An unusually early arrival of the monsoon
The June 2013 monsoon rains in Uttarakhand were highly unusual, as the monsoon came to the region two weeks earlier than normal. The monsoon started in South India near the normal June 1 arrival date, but then advanced across India in unusually rapid fashion, arriving in Pakistan along the western border of India a full month earlier than normal. Fortunately, no more heavy rain is expected in Uttarakhand over the next few days, as the monsoon will be active only in eastern India. Heavy rains are expected again in the region beginning on June 24. Wunderblogger Lee Grenci’s post, Summer Monsoon Advances Rapidly across India: Massive Flooding Ensues, has more detail on the meteorology of this year’s monsoon. There is criticism from some that the devastating floods were not entirely a natural disaster–human-caused deforestation, dam building, and mining may have contributed. “Large-scale construction of dams and absence of environmental regulations has led to the floods,” said Sunita Narian, director general of Delhi based advocacy group Centre for Science and Environment (CSE).
Figure 3. The summer monsoon arrived in southwest India right on schedule (June 1) in South India, but it spread northward much faster than usual, reaching Pakistan a full month earlier than normal. Solid green contours indicate the progress of the 2013 summer monsoon (each contour is labeled with a date). You can compare this year’s rapid advance to a “normal” progression, which is represented by the dashed, red contours (also labeled with dates).
Monsoons in India: a primer
Disastrous monsoon floods are common in India and surrounding nations, and 60,000 people–an average of 500 people per year–died in India due to monsoon floods between 1900 – 2012, according to EM-DAT, the International Disaster Database. The monsoon occurs in summer, when the sun warms up land areas more strongly than ocean areas. This happens because wind and ocean turbulence mix the ocean’s absorbed heat into a “mixed layer” approximately 50 meters deep, whereas on land, the sun’s heat penetrates at a slow rate to a limited depth. Furthermore, due to its molecular properties, water has the ability to absorb more heat than the solid materials that make up land. As a result of this summertime differential heating of land and ocean, a low pressure region featuring rising air develops over land areas. Moisture-laden ocean winds blow towards the low pressure region and are drawn upwards once over land. The rising air expands and cools, condensing its moisture into some of the heaviest rains on Earth–the monsoon. Monsoons operate via the same principle as the familiar summer afternoon sea breeze, but on a grand scale. Each summer, monsoons affect every continent on Earth except Antarctica, and are responsible for life-giving rains that sustain the lives of billions of people. In India, home for over 1.1 billion people, the monsoon provides 80% of the annual rainfall. The most deadly flooding events usually come from monsoon depressions (also known as monsoon lows.) A monsoon depression is similar to (but larger than) a tropical depression. Both are spinning storms hundreds of kilometers in diameter with sustained winds of 50 – 55 kph (30 – 35 mph), nearly calm winds at their center, and generate very heavy rains. Typically, 6 – 7 monsoon depressions form each summer over the Bay of Bengal and track westwards across India.
The future of monsoons in India
A warming climate loads the dice in favor of heavier extreme precipitation events. This occurs because more water vapor can evaporate into a warmer atmosphere, increasing the chances of record heavy downpours. In a study published in Science in 2006, Goswami et al. found that the level of heavy rainfall activity in the monsoon over India had more than doubled in the 50 years since the 1950s, leading to an increased disaster potential from heavy flooding. Moderate and weak rain events decreased during those 50 years, leaving the total amount of rain deposited by the monsoon roughly constant. The authors commented, “These findings are in tune with model projections and some observations that indicate an increase in heavy rain events and a decrease in weak events under global warming scenarios.” We should expect to see an increased number of disastrous monsoon floods in coming decades if the climate continues to warm as expected. Since the population continues to increase at a rapid rate in the region, death tolls from monsoon flooding disasters are likely to climb dramatically in coming decades. However, my greater concern for India is drought. The monsoon rains often fail during El Niño years, and more than 4.2 million people died in India due to droughts between 1900 – 2012. Up until the late 1960s, it was common for the failure of the monsoon rains to kill millions of people in India. The drought of 1965 – 1967 killed at least 1.5 million people. However, since the Green Revolution of the late 1960s–a government initiative to improve food self-sufficiency using new technology and high-yield grains–failure of the monsoon rains has not led to mass starvation in India. It is uncertain whether of not the Green Revolution can keep up with India’s booming population, and the potential that climate change might bring more severe droughts. Climate models show a wide range of possibilities for the future of the Indian monsoon, and it is unclear at present what the future might hold. However, the fact that one of the worst droughts in India’s history occurred in 2009 shows that serious droughts have to be a major concern for the future. The five worst Indian monsoons along with the rainfall deficits for the nation:
References
Goswami, et al., 2006, ” Increasing Trend of Extreme Rain Events Over India in a Warming Environment”, Science, 1 December 2006:Vol. 314. no. 5804, pp. 1442 – 1445 DOI: 10.1126/science.1132027
It was a wild weather night over much of the Midwest, Great Lakes, and Mid-Atlantic on Wednesday, as tornadoes and an organized complex of severe thunderstorms known as a bow echo brought damaging winds to a large swath of the country. NOAA’s Storm Prediction Center (SPC) logged twelve preliminary reports of tornadoes in Iowa, Illinois, and Ohio, but no injuries or major damage were reported with the twisters. A large area of severe thunderstorms organized into a curved band known as a “bow echo” over Indiana during the evening. The bow echo raced east-southeastwards at 50 mph overnight, spawning severe thunderstorm warnings along its entire track, and arrived in Washington D.C. and Baltimore, Maryland near 9 am EDT Thursday morning. SPC logged 159 reports of high thunderstorm wind gusts of 58 mph or greater in the 26 hours ending at 10 am EDT Thursday morning, and three of these gusts were 74 mph or greater. SPC did not classify this event as a “derecho”, since the winds were not strong enough to qualify. Last year’s June 29, 2012 derecho had 675 reports of wind gusts of 58 mph or greater, with 35 of these gusts 74 mph or greater. Thirteen people died in the winds, mostly from falling trees; 34 more people died from heat-related causes in the areas where 4 million people lost power in the wake of the great storm.
Another round of severe weather is expected over the Mid-Atlantic states Thursday afternoon and evening, and SPC has placed portions of this region in their “Moderate Risk” area for severe weather.
Figure 1. Lightning strikes the Willis Tower (formerly Sears Tower) in downtown on June 12, 2013 in Chicago, Illinois. (Photo by Scott Olson/Getty Images)
Figure 2. An organized line of severe thunderstorms took the shape of a “bow echo” over Indiana last night, triggering severe thunderstorm warnings along the entire front of the bow.
Figure 3. Severe weather reports for the 24 hours ending at 8 am EDT June 13, 2013, from SPC.
Big wind in the Windy City
I watched with some trepidation Wednesday evening as a large tornado vortex signature on radar developed west of Chicago, heading right for one of the most densely populated areas in the country. Fortunately, the storm pulled its punch, and Chicago was spared a direct hit by a violent tornado. But what would happen if a violent, long-track EF4 or EF5 tornado ripped through a densely populated urban area like Chicago? That was the question posed by tornado researcher Josh Wurman of the Center for Severe Weather Research in Boulder and three co-authors in a paper published in the January 2007 issue of the Bulletin of the American Meteorological Society. Their astonishing answer: damage of $40 billion and 13,000-45,000 people killed–the deadliest natural disaster in American history, eclipsing the Galveston Hurricane (8,000 fatalities.)
Figure 4. Tornadoes to affect the Chicago area, 1950-2005. Background image credit: Google Earth. Tornado paths: Dr. Perry Samson.
Huge tornado death tolls are very rare
A tornado death toll in the ten of thousands seems outlandish when one considers past history. After all, the deadliest tornado in U.S. history–the great Tri-state Tornado of March 18, 1925–killed 695 people in its deadly rampage across rural Missouri, Illinois, and Indiana. That was before the advent of Doppler radar and the National Weather Service’s excellent tornado warning system. In fact, there has only been one tornado death toll over 100 (the 158 killed in the Joplin, Missouri tornado in 2011) since 1953, the year the NWS began issuing tornado warnings. Chicago has been hit by one violent tornado. On April 21, 1967 a 200-yard wide F4 tornado formed in Palos Hills in Cook County, and tore a 16-miles long trail of destruction through Oak Lawn and the south side of Chicago. Thirty-three people died, 500 more were injured, and damage was estimated at $50 million.
Figure 5. Wind speed swaths for the 1999 F5 Mulhall, Oklahoma tornado if it were to traverse a densely populated area of Chicago. Units are in meters/sec (120 m/s = 269 mph, 102 m/s = 228 mph, and 76 m/s = 170 mph). Winds above 170 mph usually completely destroy an average house, with a crudely estimated fatality rate of 10%, according to Wurman et al.. Insets x, y, and z refer to satellite photo insets in Figure 2. Image credit: Bulletin of the American Meteorological Society.
Figure 6. Aerial photographs from Google Earth of densely populated area of Chicago (insets x, y, and z from Figure 5) These areas contain mainly single-family homes, with housing units densely packed on small lots. A mixture of three-story apartments and single-family homes is typical across the Chicago metropolitan area and many older cities such as New York City and Detroit. At lower right is a photo of Moore, OK, showing lower density housing like the 1999 Bridgecreek-Moore tornado passed through.
The paper by Wurman et al., “Low-level winds in tornadoes and the potential catastrophic tornado impacts in urban areas” opens with an analysis of the wind structure of two F5 tornadoes captured on mobile “Doppler on Wheels” radar systems–the May 3, 1999 Bridgecreek-Moore tornado, which hit the southern suburbs of Oklahoma City, and the Mulhall, Oklahoma tornado of the same day, which moved over sparsely populated rural regions. The Bridgecreek-Moore tornado had the highest winds ever measured in a tornado, 302 mph. Winds of EF4 to EF5 strength (greater than 170 mph) are capable of completely destroying a typical home, and occurred over a 350 meter (1150 foot) wide swath along this tornado’s path. The Mulhall tornado had weaker winds topping out at 245-255 mph, but had EF4 to EF5 winds over a much wider swath–1600 meters (one mile).
The F4 to F5 winds of the Bridgecreek-Moore tornado killed 36 people. Given the population of the area hit, between 1% and 3% of the people exposed to these winds died. The authors thought that this number was unusually low, given the excellent warnings and high degree of tornado awareness in Oklahoma’s population. They cited the death rate in the 1998 Spencer, South Dakota F4 tornado that destroyed 30 structures and caused six deaths, resulting in a death rate of 6% (assuming 3.3 people lived in each structure). There are no studies that relate the probability of death to the amount of damage a structure receives, and the authors estimated crudely that the death rate per totally destroyed structure is 10%. This number will go down sharply if there is a long warning time, as there was in the Oklahoma tornadoes. If one takes the Mulhall tornado’s track and superimposes it on a densely populated region of Chicago (Figure 5), one sees that a much higher number of buildings are impacted due to the density of houses. Many of these are high-rise apartment buildings that would not be totally destroyed, and the authors assume a 1% death rate in these structures. Assuming a 1% death rate in the partially destroyed high-rise apartment buildings and a 10% death rate in the homes totally destroyed along the simulated tornado’s path, one arrives at a figure of 13,000-45,000 killed in Chicago by a violent, long-track tornado. The math can applied to other cities, as well, resulting in deaths tolls as high as 14,000 in St. Louis, 22,000 in Dallas, 17,000 in Houston, 15,000 in Atlanta, and 8,000 in Oklahoma City. Indeed, the May 31, 2013 EF5 tornado that swept through El Reno, Oklahoma, killing four storm chasers, could have easily killed 1,000 people had it held together and plowed into Oklahoma City, hitting freeways jammed with people who unwisely decided to flee the storm in their cars. The authors emphasize that even if their death rate estimates are off by a factor ten, a violent tornado in Chicago could still kill 1,300-4,500 people. The authors don’t give an expected frequency for such an event, but I speculate that a violent tornado capable of killing thousands will probably occur in a major U.S. city once every few hundred years.
A historic multi-billion dollar flood disaster has killed at least eighteen people in Central Europe after record flooding unprecedented since the Middle Ages hit major rivers in Austria, the Czech Republic, Germany, Poland and Slovakia over the past two weeks. The Danube River in Passau, Germany hit the highest level since 1501, and the Saale River in Halle, Germany was the highest in its 400-year period of record. Numerous cities recorded their highest flood waters in more than a century, although in some locations the great flood of 2002 was higher. The Danube is expected to crest in Hungary’s capital city of Budapest on June 10 at the highest flood level on record, 35 cm higher than the record set in 2006. The flooding was caused by torrential rains that fell on already wet soils. In a 2-day period from May 30 – June 1, portions of Austria received the amount of rain that normally falls in two-and-half months: 150 to 200 mm (5.9 to 7.9″), with isolated regions experiencing 250 mm (9.8″). This two-day rain event had a greater than 1-in-100 year recurrence interval, according to the Austrian Meteorological Agency, ZAMG. Prior to the late May rains, Austria had its seventh wettest spring in 150 years, which had resulted in the ground in the region becoming saturated, leading to greater runoff when the rains began.
Figure 1. Aerial view of the flooded Danube River in Deggendorf, Germany on Friday, June 7, 2013. (AP Photo/Armin Wegel)
Figure 2. The Danube River in Mainz, Germany was barely kept in check by a floodwall built by IBS Engineering. Image credit: IBS Engineering.
Floods caused by a blocking high pressure system
The primary cause of the torrential rains over Central Europe during late May and early June was large loop in the jet stream that developed over Europe and got stuck in place. A “blocking high” set up over Northern Europe, forcing two low pressure systems, “Frederik” and “Günther”, to avoid Northern Europe and instead track over Central Europe. The extreme kink in the jet stream ushered in a strong southerly flow of moisture-laden air from the Mediterranean Sea over Central Europe, which met up with colder air flowing from the north due to the stuck jet stream pattern, allowing “Frederik” and “Günther” to dump 1-in-100 year rains. The stuck jet stream pattern also caused record May heat in northern Finland and surrounding regions of Russia and Sweden, where temperatures averaged an astonishing 12°C (21°F) above average for a week at the end of May. All-time May heat records–as high as 87°F–were set at stations north of the Arctic Circle in Finland.
Figure 3. Nine-day rainfall amounts in portions of Southern Germany and Western Austria exceeded 12″ (305 mm.) Image credit: ZAMG.
If it seems like getting two 1-in-100 to 1-in-500 year floods in eleven years is a bit suspicious–well, it is. Those recurrence intervals are based on weather statistics from Earth’s former climate. We are now in a new climate regime with more heat and moisture in the atmosphere, combined with altered jet stream patterns, which makes major flooding disasters more likely in certain parts of the world, like Central Europe. As I discussed in a March 2013 post, “Are atmospheric flow patterns favorable for summer extreme weather increasing?”, research published this year by scientists at the Potsdam Institute for Climate Impact Research (PIK) in German found that extreme summertime jet stream patterns had become twice as common during 2001 – 2012 compared to the previous 22 years. One of these extreme patterns occurred in August 2002, during Central Europe’s last 1-in-100 to 1-in-500 year flood. When the jet stream goes into one of these extreme configurations, it freezes in its tracks for weeks, resulting in an extended period of extreme heat or flooding, depending upon where the high-amplitude part of the jet stream lies. The scientists found that because human-caused global warming is causing the Arctic to heat up more than twice as rapidly as the rest of the planet, a unique resonance pattern capable of causing this behavior was resulting. According to German climate scientist Stefan Rahmstorf, “Planetary wave [jet stream] amplitudes have been very high in the last few weeks; we think this plays a role in the current German flooding event.” More rains are in store for the flood area through Monday, then the blocking pattern responsible for the great 2013 Central European flood is expected to disintegrate, resulting in a return to more typical June weather for the next two weeks.
Figure 4. The northward wind speed (negative values, blue on the map, indicate southward flow) at an altitude of 300 mb in the mid-latitudes of the Northern Hemisphere during July 1980, July 2011, and the last twelve days of May 2013. July of 2011 featured an unusually intense and long-lasting heat wave in the U.S. (the 4th warmest month in U.S. history), and the normally weak and irregular waves (like observed during the relatively normal July of 1980) were replaced by a strong and regular wave pattern. Late May 2013 was also very extreme, resulting the great Central European floods of 2013. Image credit: Vladimir Petoukhov and Stefan Rahmstorf.
Links Stefan Rahmstorf’s blog (translated from German) on the unusual jet stream patterns that caused the Central European floods of 2013.
NASA has high-resolution MODIS satellite images showing the flooding of the Elbe River in Germany.
Video 1. Climate, Ice, and Weather Whiplash: In this June 3, 2013 video by the Yale Climate Forum’s Peter Sinclair, Rutgers’ Jennifer Francis and Weather Underground’s Jeff Masters explore the ‘Why?’ of two years of mirror images of weather across North America.
I’m in Granby, Colorado this week for the American Geophysical Union’s Chapman Conference on Climate Change Communication. Many of the talks will be webcast live; you can see a list of the talks (times in MDT) here. My talk, “The Weather Underground Experience,” is scheduled for Monday at 3 pm MDT. I’ll give a 15-minute overview of the history of wunderground, and what I’ve learned about communicating weather and climate change information along the way. My blog updates this week may be somewhat random as a result of the conference, but I’m not seeing anything in the tropics worthy of discussion at this point
Tropical Storm Andrea is rapidly losing its tropical characteristics as it barrels northeastwards at 27 mph up the U.S. East Coast, but it still has plenty of tropical moisture that is feeding very heavy rains. Rains of 2 – 4″ are expected along a swath from South Carolina to New England from Andrea over the next two days. Pine Ridge, NC has received 6.5″ of rain from Andrea, and New River MCAS, North Carolina picked up 2″ of rain as of 9 am EDT this morning, along with a wind gust of 47 mph at 3:18 am. The same band of heavy thunderstorms spawned a possible tornado near Hubert, North Carolina at 4:45 am EDT. Andrea has spawned a preliminary count of eleven tornadoes, which is a respectable number for a landfalling June tropical storm, but not a record. According to TWC’s severe weather expert Dr. Greg Forbes, there have been two other June tropical storms since the year 2000 that spawned far more tornadoes–Tropical Storm Bill during June 29 – July 3, 2003 (32 tornadoes in FL, GA, LA, AL, MS, SC, NC, NJ), and Tropical Storm Allison of June 7 – 17 2001 (28 tornadoes in FL, AL, GA, LA, MS, SC, VA, MA, ME.) Only one of Andrea’s tornadoes caused an injury, a tornado that hit The Acreage in Palm Beach County at 6:45 am EDT. The highest storm surge from Andrea was 4.55′ at Cedar Key, Florida.
Figure 1. Yummy’s cafe in Gulfport, Florida was hit Thursday morning by a waterspout that moved ashore and became a tornado.(LAUREN CARROLL/Tampa Bay Times)
Figure 2. Predicted rainfall for the 48-hour period from 8 am EDT Friday, June 7, to 8 am EDT Sunday, June 8, 2013. Image credit: NOAA.
Video 1. NASA animation of Andrea satellite images. More cool NASA images of Andrea are here.
The Atlantic hurricane season is getting longer
Andrea’s formation in June continues a pattern of an unusually large number of early-season Atlantic named storms we’ve seen in recent years. Climatologically, June is the second quietest month of the Atlantic hurricane season, behind November. During the period 1870 – 2012, we averaged one named storm every two years in June, and 0.7 named storms per year during May and June. In the nineteen years since the current active hurricane period began in 1995, there have been fifteen June named storms (if we include 2013’s Tropical Storm Andrea.) June activity has nearly doubled since 1995, and May activity has more than doubled (there were seventeen May storms in the 75-year period 1870 – 1994, compared to 6 in the 19-year period 1995 – 2013.) Some of this difference can be attributed to observation gaps, due to the lack of satellite data before 1966. However, even during the satellite era, we have seen an increase in both early season (May – June) and late season (November – December) Atlantic tropical storms. Dr. Jim Kossin of the University of Wisconsin looked at the reasons for this in a 2008 paper titled, “Is the North Atlantic hurricane season getting longer?” He concluded that there is a “apparent tendency toward more common early- and late-season storms that correlates with warming Sea Surface Temperature but the uncertainty in these relationships is high.” He found that hurricane season for both the period 1950-2007 and 1980-2007 got longer by 5 to 10 days per decade (see my blog post on the paper.)
Invest 92L in the Central Atlantic no threat to develop Satellite images show that disorganized tropical wave is in the Central Atlantic, about a two-day journey from the Lesser Antilles Islands. NHC designated this system 92L Thursday afternoon. High wind shear of 30 – 40 knots is ripping up the thunderstorms in 92L as they form, and wind shear is predicted to remain 30 – 40 knots for the next five days, making development unlikely. The wave will likely bring heavy rain showers and gusty winds to the northern Lesser Antilles Islands beginning on Sunday night. None of the reliable computer models is showing development of a tropical cyclone in the Atlantic over the next seven days.
t was a terrifying evening of tornado chaos and extreme atmospheric violence in the Oklahoma City area on Friday. Three tornadoes touched down near the city, killing nine, injuring at least 71, and causing widespread destruction. Huge hail up to baseball-sized battered portions the the metro area, accompanied by torrential flooding rains, widespread damaging straight-line winds, and lightning that flashed nearly continuously. The strongest tornado, which touched down west of Oklahoma City in El Reno, has been preliminarily rated an EF-3 with 136 – 165 mph winds. The tornado warning for the storm was issued 19 minutes before it touched down. Two other EF-3 tornadoes touched down near St. Louis, Missouri, and NOAA’s Storm Prediction Center (SPC) logged 20 preliminary tornado reports on Friday. Tinker Air Force Base on the east side of Oklahoma City reported sustained winds of 68 mph, gusting to 88 mph, at 8:09 pm CDT. The Oklahoma City airport had sustained winds of 53 mph, gusting to 71 mph at 7:26 pm. These winds were generated by the massive and powerful downdrafts from the supercell thunderstorm that spawned the El Reno tornado. Thankfully, Friday was likely the peak day for this week’s severe weather outbreak, as SPC is calling for only a “Slight Risk” of severe weather Saturday and Sunday.
Figure 1. TWC’s Mike Bettes crew caught this image of the El Reno, Oklahoma tornado of May 31, 2013 before the tornado caught them and rolled their vehicle.
Figure 2 and 3. Radar reflectivity (top) and Doppler velocity (bottom) images of the May 31, 2013 El Reno, Oklahoma tornado.
Figure 4. Preliminary tracks of the three tornadoes that touched done near Oklahoma City on May 31, 2013. Image credit: NWS Norman, OK.
Tornadoes and cars: a dangerous mix
A vehicle is about the worst place you can be in a tornado, as the tornado’s winds can easily roll a car. (The only place less safe is probably a mobile home, as a tornado’s winds can roll mobile homes almost as readily, and mobile homes don’t come with seat belts and air bags.) At least five of the deaths in Friday’s El Reno tornado occurred in vehicles attempting to flee. There was one local TV station that urged residents without underground shelters to get in their cars and “get south” in advance of the tornado that was approaching Oklahoma City, since chasers were reporting that the El Reno tornado may have been so strong that only an underground shelter would have provided adequate protection. This terrible piece of advice likely contributed to the incredible traffic jams that we saw on I-35, I-40, I-44, and other local roads Friday night. Thousands of cars were bumper-to-bumper on the roads as a dangerous tornado approached them. Had the El Reno tornado plowed directly down one of these car-choked interstates, the death toll could have easily exceeded 500. If you are located in a metro area and don’t have an underground shelter, the best thing to do it to take shelter in an interior windowless room or hallway, with protective furniture over your body. Getting in a car and attempting to flee the tornado is the worst thing you can do in an urban area. You may not be able to see the tornado if it is dark or the tornado is wrapped in rain. You are likely to encounter hazardous winds, rain, and hail, run into unexpected traffic, or flooded or debris-blocked roads that will put you directly in the path of the tornado. Even without an underground shelter, most people will be able to survive a dangerous EF-4 tornado. Case in point: during the Mannsford, Oklahoma EF- 4 tornado of 1984, a packed church received a direct hit, and everyone in the church survived. The only fatality was a man who drove to the church to get his wife. (Thanks to wunderground member AGWcreationists for this link.) It’s better to abandon your vehicle and take shelter in a ditch, if you are caught in a car during a tornado.
Video 1. The Weather Channel storm chasers weren’t the only ones who got themselves in an extremely dangerous situation on May 31. StormChasingVideo.com storm chaser Brandon Sullivan and his chase partner Brett Wright got caught in the tornado northwest of Union City, OK and slammed with debris as the tornado hit a barn that exploded in front of them.
Video 2. When the hunters became the hunted: Weather Channel storm chasers Mike Bettes and two photographers were in their Tornado Hunt vehicle when they were hit by a tornado in El Reno, Oklahoma on May 31, 2013. The tornado picked their car up off the ground and rolled it 6 – 8 times before depositing it in a field 200 yards away. All the occupants were wearing seat belts and the air bags deployed, likely saving their lives. Bettes sustained minor injuries, including stitches in his hand. It was the first injury sustained by a Weather Channel personality covering violent weather, according to company spokesperson Shirley Powell.
A storm chasers’ nightmare
Cars and tornadoes can prove a dangerous mix even for the world’s most experienced storm chasers. Driving at high speeds though heavy rain, large hail, and high winds is hazardous. If one is lucky enough to chase down a tornado, even the most experienced chasers can find themselves in a serious life-threatening situation when unpredictable events occur. Tornadoes by their nature are unpredictable, and can change course unexpectedly, or pop up suddenly. It’s particularly dangerous when a tornado is wrapped in rain, making it hard to see, or if a chaser is operating in a heavily populated area, where roads may suddenly become congested. All four of these conditions occurred Friday during the El Reno tornado, and it is very fortunate that multiple chasers were not killed. The El Reno tornado was wrapped in rain and difficult to see as it headed west towards Oklahoma City. The twister suddenly made a jog to the southeast as a Weather Channel team led by Mike Bettes was attempting to get in front of the storm, and the tornado lifted their vehicle off the ground, rolled it multiple times, and hurled it 200 yards into a nearby field. StormChasingVideo.com storm chaser Brandon Sullivan and his chase partner Brett Wright got caught in the tornado northwest of Union City, OK and slammed with debris as the tornado hit a barn that exploded in front of them. Meteorologist Emily Sutton and storm chaser Kevin Josefy of local Oklahoma City TV station KFOR also had a very close call with the El Reno tornado Friday afternoon. They got too close to the tornado, and were forced to floor the car in reverse to escape flying debris. With branches of trees crashing around them, Sutton began feeling debris hitting her back, and realized that the rear windshield of the car must have gotten destroyed. Both were uninjured. Reed Timmer’s armor-plated “Dominator” chase vehicle had its hood torn off by the tornado. Wunderground member Levi32 was out storm chasing during the El Reno Tornado, and got stuck in traffic on Highway 4 and couldn’t move. “We looked up above the car and saw the wall cloud over top of us, with very quick rotation and rising scud indicating the updraft. We were definitely too close. We made it home safely last night, but not until after an insanely wild day. One hour of chasing turned into six more of being chased by at least 2 tornadoes and a 3rd wall cloud, one of which was the one that went right through downtown Oklahoma City. At one point we were stuck in traffic underneath the El Rino wall cloud watching rotating, rising scud directly above the car. I am hoping and praying that the daylight does not reveal more fatalities.
Would I go again? Yes, but not today, or tomorrow, and I would take even greater care. We had no clue we would get caught the way we did. I thought we had done everything right. We were kind of freaking out for a while. That velocity signature you guys saw with radar folding and multiple vortices – we were under the southern edge of it. We never got a clear view of the tornado, but we could tell just how close it was to our north. It was unreal. The inflow got pretty strong.
We were almost ready to jump out and take cover right before we found a route south, which ended up being slow. It became a six-lane highway south as everyone panicked and drove on the wrong side of the road. Even we did so. We thought we were clear until we saw the training of tornadic supercells on radar, all connected somehow. I’ve never seen anything like that. My best pictures of the day were of the wall cloud that followed behind the El Reno storm. We didn’t see a funnel from that one either, but it chased us south for a long time, and we heard from radio that it spawned a confirmed tornado in Tuttle, when we realized that we were in Tuttle.
A third mesocyclone showed up behind that one as we continued slowly south, eventually reaching Blanchard. It looked weaker than the others but we weren’t going to escape it, so we took shelter in a storm room in the local grocery store for about an hour. It then took a long time to find a way around the huge hail cores to get back home. Lightning flashes were occurring 10 times per second as we drove home in the dark. It was almost calming to watch as we got over the semi-shock that we were all in. None of us in the car had seen a tornado before. We didn’t see one yesterday, but we were chased by two.”
Video 3. Birth of the El Reno wedge tornado. As the tornado touched down, it produced a rare display of suction vortices.
Video 4. Storm chasers Jeff Piotrowski and Kathryn Piotrowski captured impressive footage of a double vortex tornado near El Reno, Oklahoma on May 31, 2013.
Severe storms causing major flooding
The 5.64″ of rain that fell at the Oklahoma City Will Rogers Airport on Friday was their 6th wettest day in city history, and brought the total rainfall for the month of May to 14.52″, the wettest May in Oklahoma City’s history (Thanks to BaltimoreBrian for this link.) The North Canadian River in Oklahoma City rose sixteen feet in twelve hours, cresting at its 2nd highest flood on record this Saturday morning. The heavy rains have spread eastwards on Saturday, causing more flooding problems. Paducah, KY had its wettest June day and 3rd wettest day on record on June 1, with 5.73″ of rain (all-time record: 7.49″ on 9/5/1985.) Major flooding is occurring along a substantial stretch of the Mississippi River in Iowa, Illinois, and Missouri.
Figure 5. The North Canadian River in Oklahoma City rose sixteen feet in twelve hours, reaching its 2nd highest flood on record this Saturday morning.
Figure 6. Radar-estimated rainfall in the Oklahoma City area reached 8+” over some areas from Friday’s storm.
Remains of Hurricane Barbara may bring heavy rains to Mexico, Florida, and Cuba
Today, June 1, is the official first day of the Atlantic Hurricane season, and we already have our first Atlantic tropical disturbance to talk about. Hurricane Barbara, which died on Thursday as it attempted to cross Mexico’s Isthmus of Tehuantepec into the southernmost Gulf of Mexico, has left behind an area of disturbed weather over the southernmost Gulf of Mexico. There is very little heavy thunderstorm activity associated with Barbara’s remnants apparent on satellite loops this Saturday afternoon. Wind shear is a high 20 knots in the region, and the area of disturbed weather is quite small, so I don’t expect any development to occur over the next few days. NHC is giving the disturbance a 10% chance of developing into a tropical cyclone by Monday. Moisture from the remnants of Barbara may combine with moisture from an area of heavy thunderstorms expected to build over the Western Caribbean this weekend, and begin bringing heavy rains to Mexico’s Yucatan Peninsula and Western Cuba on Sunday and Monday. These heavy rains may spread to Southwest Florida as early as Monday night. The computer models predict that this disturbance should be large and poorly organized, making development into a Gulf of Mexico tropical cyclone unlikely.
NOAA forecasts an above-normal and possibly very active Atlantic hurricane season in 2013, in their May 23 outlook. They give a 70% chance of an above-normal season, a 25% chance of an near-normal season, and 5% chance of a below-normal season. They predict a 70% chance that there will be 13 – 20 named storms, 7 – 11 hurricanes, and 3 – 6 major hurricanes, with an Accumulated Cyclone Energy (ACE) 120% – 205% of the median. If we take the midpoint of these numbers, NOAA is calling for 16.5 named storms, 9 hurricanes, 4.5 major hurricanes, and an ACE index 162% of normal. This is well above the 1981 – 2010 average of 12 named storms, 6 hurricanes, and 3 major hurricanes. Hurricane seasons during the active hurricane period 1995 – 2012 have averaged 15 named storms, 8 hurricanes, and 4 major hurricanes, with an ACE index 151% of the median. Only five seasons since the active hurricane period that began in 1995 have not been above normal–including four El Niño years (1997, 2002, 2006, and 2009), and the neutral 2007 season.
Figure 1.Hurricane Michael as seen by NASA’s Aqua satellite at 12:20 pm EDT Thursday September 6, 2012. At the time, Michael was a major Category 3 hurricane with 115 mph winds. Hurricane Sandy was the only other major Atlantic hurricane of 2012. Image credit: NASA.
The forecasters cited the following main factors that will influence the coming season:
1) Above-average sea surface temperatures (SSTs) are expected in the hurricane Main Development Region (MDR), from the Caribbean to the coast of Africa between between 10°N and 20°N. SSTs in the MDR during April were 0.4°C above average, and were 0.33°C above the oceans in the remainder of the global tropics. Long-range seasonal computer model forecasts predict a continuation of above-average SSTs in the MDR during much of hurricane season.
2) We are in an active period of hurricane activity that began in 1995, thanks to a natural decades-long cycle in hurricane activity called the Atlantic Multi-decadal Oscillation (AMO).
3) No El Niño event is expected this year. El Niño events tend to suppress Atlantic hurricane activity. Neutral conditions have been present since last summer, and are predicted to remain neutral through hurricane season by most of the El Niño computer forecast models.
NOAA said, “This combination of climate factors historically produces above-normal Atlantic hurricane seasons. The 2013 hurricane season could see activity comparable to some of the very active seasons since 1995.” NOAA is increasingly using output from ultra-long range runs of the computer forecast models we rely on to make day-to-day weather forecasts, for their seasonal hurricane forecasts. These models include the NOAA Climate Forecast System (CFS), NOAA Geophysical Fluid Dynamics Lab (GFDL) model CM2.1, the European Centre for Medium Range Weather Forecasting (ECMWF) model, the United Kingdom Meteorology (UKMET) office model, and the EUROpean Seasonal to Inter-annual Prediction (EUROSIP) ensemble.
How accurate are NOAA’s seasonal hurricane forecasts? A talk presented by NHC’s Eric Blake at the 2010 29th Annual AMS Conference on Hurricanes and Tropical Meteorology studied the accuracy of NOAA’s late May seasonal Atlantic hurricane forecasts, using the mid-point of the range given for the number of named storms, hurricanes, intense hurricanes, and ACE index. Over the past twelve years, a forecast made using climatology was in error, on average, by 3.6 named storms, 2.5 hurricanes, and 1.7 intense hurricanes. NOAA’s May forecast was not significantly better than climatology for these quantities, with average errors of 3.5 named storms, 2.3 hurricanes, and 1.4 intense hurricanes. Only NOAA’s May ACE forecast was significantly better than climatology, averaging 58 ACE units off, compared to the 74 for climatology. Using another way to measure skill, the Mean Squared Error, May NOAA forecasts for named storms, hurricanes, and intense hurricanes had a skill of between 5% and 21% over a climatology forecast. Not surprisingly, NOAA’s August forecasts were much better than the May forecasts, and did significantly better than a climatology forecast.
Figure 3. Forecast skill of the TSR, NOAA (National Oceanic and Atmospheric Administration) and CSU (Colorado State University) for the number of hurricanes in the Atlantic during 2003-2012, as a function of lead time. Forecast precision is assessed using the Mean Square Skill Score (MSSS) which is the percentage improvement in mean square error over a climatology forecast (six hurricanes in a given year.) Positive skill indicates that the model performs better than climatology, while a negative skill indicates that it performs worse than climatology. Two different climatologies are used: a fixed 50-year (1950-1999) climatology, and a running prior 10-year climate norm. NOAA does not release seasonal outlooks before late May, and CSU stopped providing quantitative extended-range December hurricane outlooks in 2011. Skill climbs as the hurricane season approaches, with modest skill levels by early June, and good skill levels by early August. Image credit: Tropical Storm Risk, Inc (TSR).
TSR predicts an active hurricane season: 15.3 named storms
The May 24 forecast for the 2013 Atlantic hurricane season made by British private forecasting firm Tropical Storm Risk, Inc. (TSR) calls for an active season with 15.3 named storms, 7.5 hurricanes, 3.4 intense hurricanes, and an Accumulated Cyclone Energy (ACE) of 130. The long-term averages for the past 63 years are 11 named storms, 6 hurricanes, 3 intense hurricanes, and an ACE of 103. TSR rates their skill level as modest for these late May forecasts–11% – 25% higher than a “no-skill” forecast made using climatology. TSR predicts a 63% chance that U.S. land falling activity will be above average, a 21% chance it will be near average, and a 16% chance it will be below average. They project that 4.4 named storms will hit the U.S., with 2 of these being hurricanes. The averages from the 1950-2012 climatology are 3.1 named storms and 1.4 hurricanes. They rate their skill at making these late May forecasts for U.S. landfalls just 8% – 12% higher than a “no-skill” forecast made using climatology. In the Lesser Antilles Islands of the Caribbean, TSR projects 1.5 named storms, 0.6 of these being hurricanes. Climatology is 1.1 named storms and 0.5 hurricanes.
TSR’s two predictors for their statistical model are the forecast July – September trade wind speed over the Caribbean and tropical North Atlantic, and the forecast August – September 2013 sea surface temperatures in the tropical North Atlantic. Their model is calling for warmer than average SSTs and slower than average trade winds during these periods, and both of these factors should act to increase hurricane and tropical storm activity.
UKMET office predicts a slightly above normal Atlantic hurricane season: 14 named storms
The UKMET office forecast for the 2013 Atlantic hurricane season, issued May 13, calls for slightly above normal activity, with 14 named storms, 9 hurricanes, and an ACE index of 130. In contrast to the statistical models relied upon by CSU, TSR, and NOAA, the UKMET forecast is done strictly using two dynamical global seasonal prediction systems: the Met Office GloSea5 system and ECMWF system 4. In 2012, the Met Office forecast was for 10 tropical storms and an ACE index of 90. The actual numbers were 19 named storms and an ACE index of 123.
WSI predicts an active hurricane season: 16 named storms
The April 8 forecast from the private weather firm WSI (part of The Weather Company, along with The Weather Channel, Weather Central, and The Weather Underground), is calling for an active season with 16 named storms, 9 hurricanes, and 5 intense hurricanes.
Penn State predicts an active hurricane season: 16 named storms
The May 11 forecast made using a statistical model by Penn State’s Michael Mann and alumnus Michael Kozar is calling for an active Atlantic hurricane season with 16 named storms, plus or minus 4 storms. Their prediction was made using statistics of how past hurricane seasons have behaved in response to sea surface temperatures (SSTs), the El Niño/La Niña oscillation, the North Atlantic Oscillation (NAO), and other factors. The statistic model assumes that in 2013 the May 0.87°C above average temperatures in the MDR will persist throughout hurricane season, the El Niño phase will be neutral, and the North Atlantic Oscillation (NAO) will be near average.
The PSU team has been making Atlantic hurricane season forecasts since 2007, and these predictions have done pretty well, except for in 2012, when an expected El Niño did not materialize:
2007 prediction: 15 named storms, Actual: 15
2009 prediction: 12.5, named storms, Actual: 9
2010 prediction: 23 named storms, Actual: 19
2011 prediction: 16 named storms, Actual: 19
2012 prediction: 10.5 named storms, Actual: 19
The wunderground community predicts an active hurricane season: 17 named storms
Over 100 members of the wunderground community have submitted their seasonal hurricane forecasts, which are compiled on trHUrrIXC5MMX’s blog. The April 28 version of this list called for an average of 17 named storms, 8 hurricanes, and 4 intense hurricanes in the Atlantic. This list will be updated by June 3, so get your forecasts in by then! As usual, I am abstaining from making a hurricane season forecast. I figure there’s no sense making a forecast that will be wrong nearly half the time; I prefer to stick to higher-probability forecasts.
NOAA predicts a below-average Eastern Pacific hurricane season: 13.5 named storms NOAA’s pre-season prediction for the Eastern Pacific hurricane season, issued on May 23, calls for a below-average season, with 11 – 16 named storms, 5 – 8 hurricanes, 1 – 4 major hurricanes, and an ACE index 60% – 105% of the median. The mid-point of these ranges gives us a forecast for 13.5 named storms, 6.5 hurricanes, and 2.5 major hurricanes, with an ACE index 82% of average. The 1981 – 2010 averages for the Eastern Pacific hurricane season are 15 named storms, 8 hurricanes, and 4 major hurricanes. So far in 2013, there has already been one named storm. On average, the 2nd storm of the year doesn’t form until June 25.
NOAA predicts a below-average Central Pacific hurricane season: 2 tropical cyclones NOAA’s pre-season prediction for the Central Pacific hurricane season, issued on May 22, calls for a below-average season, with 1 – 3 tropical cyclones. An average season has 4 – 5 tropical cyclones, which include tropical depressions, tropical storms, and hurricanes. Hawaii is the primary land area affected by Central Pacific tropical cyclones.
The week ahead: 91E, and a heavy rainfall threat to Mexico
We’re already behind last year’s pace for named storms in both the Atlantic (where Tropical Storm Alberto formed on May 19, and Tropical Storm Beryl on May 26), and in the Eastern Pacific, where Bud formed on May 21 (the earliest date since record keeping began in 1949 for formation of the season’s second named storm.) The Madden Julian Oscillation (MJO), a pattern of increased thunderstorm activity near the Equator that moves around the globe in 30 – 60 days, is currently located in the Eastern Pacific. The MJO is relatively weak, but is helping boost the chances that Invest 91E in the Eastern Pacific will develop. On Friday, NHC was giving 91E a 20% of developing into a tropical cyclone by Sunday. The 12Z Friday runs of the GFS and ECMWF models were predicting that a weak circulation off the coast of Costa Rica, well east of the separate circulation currently called 91E, could develop into a tropical depression by Tuesday. This system is a threat to spread heavy rains to the coast of Mexico from Acapulco to Guatemala on Tuesday and Wednesday.
In the Atlantic, the models are depicting high wind shear through June 1 over the majority of the regions we typically see May tropical cyclone development–the Caribbean, Gulf of Mexico, and Bahamas. The GFS model is showing a decrease in wind shear over the Western Caribbean after June 1, which would argue for an increased chance of tropical storm development then (though wind shear forecasts more than 7 days in advance are highly unreliable.) The prospects for an early June named storm in the Atlantic are probably above average, though, given that the MJO may be active in the Atlantic during th first week of June.
