Category 6™

The Rains of May and the Science of Recurrence Intervals

By: Bob Henson , 3:54 PM GMT on May 30, 2015



As this memorable month sloshes toward its final weekend, rainfall records have continued to accumulate across the Southern Plains. May is already the wettest month on record in the statewide averages for both Oklahoma (14.43” as of 9:30 am CDT May 30, crushing the 10.75” from October 1941) and Texas (7.54” as of May 27, beating out the ominous 6.66” from June 2004). This is a startling feat when you consider that the two states’ combined area is almost the size of Egypt. Several individual locations have also scored their wettest months on record, including Wichita Falls, TX; Oklahoma City, OK; and Colorado Springs, CO (more on these and other local records next week). The onslaught of wetness is now winding down just as May comes to a close. On the plus side, Oklahoma state climatologist Gary McManus has dubbed this wet period “the death of the Southern Plains drought,” one that had dug its claws into the region for most of the 2010s. However, the associated flooding has killed at least 25 people, with 14 missing, and damaged or destroyed many hundreds of structures.
 

Figure 1. Motorists are stranded along I-45 in north Houston on Tuesday, May 26, after overnight storms flooded the area. Image credit: Cody Duty/Houston Chronicle via AP.
 
 
When an event like this happens, it’s human nature to wonder just how unusual it is. Climate change has added urgency to such questions, but even in a constant (“stationary”) climate, precise answers on the rarity of heavy rain at a given location can be hard to come by. One of the most common strategies is to calculate the “average recurrence interval” (ARI), or the typical amount of time that would separate events of a certain magnitude when considered over a much longer time period. At heart, this is simply a different name for what was long called the return period—i.e., a “100-year flood.” The problem with the “XXX-year flood” wording is that it’s wide-open to misinterpretation. People may think that a 100-year flood won’t be followed by a similarly sized event until a century has passed, when in fact it could recur any time. (You might also describe this as a “1% probability flood,” meaning there’s a 1% chance of getting it in a given year.)



NOAA has decades of return-period estimates under its belt. From the 1950s into the 2000s, these were published as hard-copy maps (see PDFs). More recently, NOAA has undertaken a comprehensive effort called Atlas 14 that is updating the return periods (now called ARIs) and putting them into a fully digital, gridded, GIS-compatible dataset. Since we’re now well into the 21st century, the project has more than a hundred years of rain-gauge measurements to draw on, from thousands of U.S. sites. The Atlas 14 results for Texas won’t be ready until at least 2018, but the Oklahoma numbers are already online, and they provide an useful window on the rarity of the rains that befell the Sooner State this month.






Figure 2. 30-day rainfall totals for Oklahoma, valid from 10:00 am CDT April 30 through May 30. Most of the heavy rain occurred from May 5 onward. Image credit: Oklahoma Mesonet.
 
 
Take long-suffering Norman, OK, which has racked up one of the state’s highest totals for the past 30 days: 23.39” as of 10:00 am CDT Saturday, May 30. (Even higher totals can be found in southeast Oklahoma, but that region also gets more rain in a typical year, so its ARIs for this event aren’t as impressive.) For a 30-day period in Norman, the best estimate of ARIs from Atlas 14 is 500 years for a rainfall of 22.4,” and 1000 years for 24.0.” When a 90% confidence interval is employed, the 500-year range is 16.3” to 29.2”, and the 1000-year range is 16.9” to 32.0”. No matter how you slice it, this goes well beyond a once-in-a-lifetime event for Normanites when considered for the month as a whole. The 24-hour numbers for Norman are less spectacular: the biggest daily rain of the month so far, 4.67” on May 19, falls very near the 5-year ARI value of 4.66”. So it’s the full month of rain that’s much more noteworthy.
 
 

Figure 3. Maximum 24-hour rainfall totals for the Houston area during the period from 1200 GMT June 5 to 11, 2001 (top), as Tropical Storm/Depression Allison dawdled over eastern Texas, and preliminary 24-hour totals from 1200 GMT May 25 to 26, 2015 (bottom), the day that brought massive flooding to western parts of the Houston metro area. The two maps are at the same scale. The analyses are based on data from more than 450 raingauges. Image credit: Applied Weather Associates and MetStat.
 




Even though Atlas 14 data aren’t yet available for Texas, we can gain some ARI-based insight on this storm thanks to the work of Tye Parzybok. His company, MetStat, has calculated new ARIs for 6- and 24-hour rainfall under a subcontract with Applied Weather Associates for the Texas Water Development Board. MetStat produces near-real-time rainfall estimates and predictions and ARI maps for clients and the public, drawing on radar data as well as more than 20,000 North American rain guages, including many compiled through the Citizen Weather Observer Program. (See the MetStat Extreme Precipitation Blog  for writeups on recent events.)
 
Examining the rainfall in and around Houston on Monday, May 25, MetStat found 24-hour ARIs of close to 1000 years for parts of the southwest Houston area (see Figure 4 below), where the gridded analysis indicates a 24-hour maximum of 13.48”. The water then flowed downstream to produce major flooding in central Houston, where the actual rainfall was much less. During Tropical Storm Allison in June 2001, extremely heavy rains covered a much larger part of southeast Texas, with several days of downpours encompassing some truly extreme totals (see Figure 3 above). One rain gauge at Greens Bayou and Mount Houston Parkway measured a 6-hour total of 20.64” during Allison, and the MetStat gridded analysis suggests a nearby maximum as high as 20.98”.




Figure 4. A preliminary map of average recurrence intervals (ARIs) for the 24-hour rains observed in and around Houston from 7:00 am CDT May 25 to 7:00 am May 26, 2015. The map was produced using 820 rain-gauge reports and the MetStorm analysis system. The highest ARI at a given point was 932 years. Image credit: MetStat, Weather Decision Technologies, and Synoptic.




It’s the weeks-long persistence that seems to be most unusual about this month’s rain in both Texas and Oklahoma. There’s an obvious parallel with this past February, which was marked by an unprecedented month-long barrage of cold and snow in much of New England and record winter warmth in and near California. These events raise questions (beyond the scope of this blog post!) about why the steering flow across North America this year has been so prone to multiweek blocking episodes. 



And now, the caveats


ARIs play a huge role in public policy. As Parzybok notes in an EarthZine article, urban water systems are widely designed for ARI rainfalls of 1 to 10 years, while highways are generally built to reflect 10- to 25-year ARIs. While policymakers and the public need and want some kind of yardstick for extreme rain events—and ARIs are the best one we have—they’re also an imperfect tool and shouldn’t be taken as gospel. Some reasons:



It’s not surprising to have rare events over small areas. Very localized extremes may have very long recurrence intervals at a specific point. If you live in Oklahoma, the ARI for a major tornado at your house might be hundreds of years, yet dozens of twisters occur across the state in a typical spring.

Rainfall can’t be measured everywhere at once.  Outside of urban areas, century-long observation points may be separated by tens of miles, which means some huge but localized downpours may be omitted from the long-term record.



Rare events are hard to contextualize, precisely because they’re so rare.  Scientists can use certain statistical techniques to calculate ARIs of up to 200 years from a regional cluster of century-long raingauge records. Beyond that threshold, there’s a lot more uncertainty in the ARIs, as illustrated by the example above from Norman.

Our climate isn’t stationary.  Climate science tells us that a warming planet will intensify rains as well as drought impacts, due largely to the increased evaporation of water from both oceans and land. More than 15 years of observational study backs this up, showing that the heaviest precipitation events are indeed becoming more intense in many parts of the world.
 
A new analysis released on Wednesday by Climate Central highlights the increasing risk of one-day downpours, defined as the wettest 1% of all wet days at a given spot for the period 1950 through 2014. Since precipitation is so variable in time and space, the state-by-state analysis is a patchwork. Most states show increases (as high as 104% in Rhode Island), while six states show minor declines (the largest being -6% in California). Overall, if we assume that downpours at various time scales are becoming more common in most parts of the nation, then it would follow that ARIs from Atlas 14 will tend to be on the high side—i.e., in many locations, we would expect an event diagnosed with a 100-year ARI to occur a bit more often as climate change proceeds.
 
The interactive below shows how the frequency of heavy downpours is evolving over time across the nation. For more background, see the accompanying research report from Climate Central (the interactive is also available at that report site].
 
Bob Henson
 

 

Figure 5. Heavy rains soaked parts of the drought-stricken Northeast on Thursday, May 28, including Portland, Maine. Image credit: wunderphotographer Alphaholik.

Flood Extreme Weather

Earth's 5th Deadliest Heat Wave in Recorded History Kills 1,826 in India

By: Jeff Masters , 12:34 PM GMT on May 29, 2015

The death toll from India's horrid May heat wave has risen to 1,826, making this year's heat wave the second deadliest in India's recorded history--and the fifth deadliest in world history. According to statistics from EM-DAT, the International Disaster Database, India's only deadlier heat wave was in 1998, when 2,541 died. With over 400 deaths recorded in just the past day and the heat expected to continue over India for another week, the 1998 death toll could well be exceeded in this year's heat wave. However, death tolls from heat waves are very difficult to estimate, since excess heat is typically not listed as the primary cause of death in cases where the victim has a pre-existing condition such as heart or lung disease. For example, the U.S. National Climatic Data Center (NCDC) lists the total direct deaths from the U.S. heat wave of 1980 at 1,260, but estimates that the combined direct and indirect deaths (i.e., excess mortality) due to heat stress was 10,000. Below is the list of top ten deadliest heat waves in world history as compiled by EM-DAT, the International Disaster Database, which uses direct deaths for their statistics, and not excess mortality.

The 10 Deadliest Heat Waves in World History
1) Europe, 2003: 71,310
2) Russia, 2010: 55,736
3) Europe, 2006: 3,418
4) India, 1998: 2,541
5) India, 2015: 1,826+
6) U.S. and Canada, 1936: 1,693
7) U.S., 1980: 1,260
8) India, 2003: 1,210
9) India, 2002: 1,030
9) Greece and Turkey, 1987: 1,030

Note that the EM-DAT database may not be entirely reliable; for example, they list no heat deaths in the U.S. for the 1988 heat wave, while the U.S. National Climatic Data Center (NCDC) lists 454 direct deaths and 5,000 combined direct and indirect deaths. The 2010 Japanese heat wave, which EM-DAT gives a death toll of 170 for, disagrees with the 1,718 total from the Ministry of Health, Labour and Welfare of Japan. Weather records researcher Maximiliano Herrera alleges that the deadliest and most brutal heat wave Chinese history, in Eastern China in the summer 2013, had thousands of deaths which were not reported by the Chinese authorities. The official death toll was merely 40.


Figure 1. A young Indian child pours water on himself as he tries to cool himself off in New Delhi on May 28, 2015. Image credit: MONEY SHARMA/AFP/Getty Images.

It's the heat and the humidity
Temperatures across much of India have been 5°C (9°F) above average this May, with very high humidity. In many of the hardest-hit areas of eastern India, the heat index dropped below 100°F for only four hours each night for several consecutive days this week. This sort of day-after-day heat stress is very hard on vulnerable people, and leads to high mortality. For example, in Channai (Madras) on May 24, the high temperature reached 108°F and the heat index topped out at 123°F, and never dropped below 97°F the entire day. Far more extreme heat index values have been observed in some areas. For example, on May 23 at 14:30,  Bhubneshwar recorded a temperature of of 42.2°C (108°F) with a dew point of 29.3°C (84.7°F), giving an astonishing heat index of 62°C (143.6°F.) According to weather records researcher Maximiliano Herrera, a heat index of up to 65°C (149°F) has been measured at some stations in eastern India during the heat wave.


Figure 2. Progress of the monsoon towards India as of May 28, 2015 (green line) has been close to its average pace. If the monsoon follows its usual pace, it will move through the province hardest hit by this year's heat wave, Andhra Prahesh (shaded in yellow), by June 5. This province recorded 1,334 heat deaths as of May 29, 2015. Image credit: India Meteorological Department .

The monsoon is coming
This is the time of year when India's 1.2 billion people look beseechingly southwards, toward the advancing southwest monsoon. The monsoon's arrival brings rains that cool India's scorching May heat, and the monsoon's rains give life, providing 70 - 80% of the year’s total rainfall in just four months. The India Meteorological Department (IMD) is forecasting that the arrival of the southwest monsoon at the southern tip of India will occur this weekend, on May 30. This is two days ahead of the average arrival date, June 1. The monsoon should move through the province hardest hit by this year's heat wave, Andhra Prahesh, by June 5. However, IMD is also forecasting a roughly doubled chance of below-average rains during the summer monsoon period, and predicts only 91 percent of the usual rainfall will occur. The problem: the atmospheric circulation patterns brought on by an El Niño event usually cause much reduced monsoon rains. The current borderline weak/moderate El Niño event is forecast to intensify this summer, and this is likely to cause a significant reduction in monsoon rainfall over India. According to EM-DAT, the International Disaster Database, more than 4.2 million people died in India between 1900 - 2014 due to droughts from failed monsoon rains, primarily during El Niño years. The five worst Indian monsoons for rainfall deficit:

1) 1877, -33%
2) 1899, -29%
3) 1918, -25%
4) 1972, -24%
5) 2009, -22%

Up until the late 1960s, it was common for the failure of the monsoon rains to kill millions of people in India; the 3-year drought that began during the strong El Niño event of 1965 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 famine in India. For example, the fifth worst drought in India's history occurred in 2009, but did not result in serious food shortages--and neither would a similar failure of the monsoon this year. However, a weak monsoon could affect India’s fragile power supply, since the country is heavily dependent on hydropower. In 2012, a weak monsoon forced farmers to use huge amounts of power to pump groundwater to make up for lack of rain. The resulting strain on the power grid helped trigger a blackout that affected 600 million people. Fortunately, many reservoirs in India are above their 10-year average level heading into the summer.

Climate change and India
This year's deadly heat wave in India was made much more probable by the fact that Earth is experiencing its hottest temperatures on record--the past twelve months were the warmest twelve-month period in recorded history, and so was the January - April 2015 period. According to the India Meteorological Department, a warming climate increased heat waves in India by a third between 1961 to 2010. As the planet continues to warm due to human-caused global warming, heat waves will become more frequent and more intense, and heat-related deaths will soar unless we take strong measures to adapt. An April 2015 paper published in Regional Environmental Change, Intensification of future severe heat waves in India and their effect on heat stress and mortality, warned that "heat waves are projected to be more intense, have longer durations and occur at a higher frequency and earlier in the year. Southern India, currently not influenced by heat waves, is expected to be severely affected by the end of the twenty-first century." Perhaps a bigger concern for India with climate change is drought, though. Many climate models show that climate change might increase the average rainfall in India from the monsoon, but when dry years occur, the hotter temperatures accompanying the dry years will drive much more intense droughts capable of causing significant challenges to growing food in India.

Links
The May 27, 2015 post by Eric Holthaus of Slate discusses the India heat wave and climate change.

Wunderground's climate change blogger Dr. Ricky Rood wrote a nice 3-part series about the challenges India faces due to climate change after he completed a 2009 trip there.

Bob Henson will have a new post on the Texas/Oklahoma flood situation on Friday evening or Saturday morning.

Jeff Masters

Heat Climate Change

Tropical Storm Andres Forms in the Northeast Pacific; Not a Threat to Mexico

By: Jeff Masters , 2:43 PM GMT on May 28, 2015

The Northeast Pacific's first named storm of 2015 is here. Tropical Storm Andres formed at 11 am EDT on Thursday, in the waters about 690 miles southwest of Manzanillo, Mexico. The first named storm of the Northeast Pacific hurricane season usually forms by June 10, so we are nearly two weeks ahead of climatology. According to the database of Eastern Pacific storms maintained by NOAA's Office for Coastal Management, the formation of a tropical storm in May in the Eastern Pacific is not uncommon, though--33 named storms have formed in May in the 45 years since accurate satellite records began in 1970. Sea surface temperatures (SSTs) are unusually warm in the region, at least 2°F (1.1°C) above average, thanks in large part to the intensifying El Niño event underway in the Eastern Pacific. Andres will likely be able to take advantage of these warm waters and become a hurricane this weekend, but the storm will stay well out to sea and not affect any land areas for at least the next five days.


Figure 1. Latest satellite image of Andres.


Figure 2. Tracks of all May tropical storms and hurricanes in the Eastern Pacific Ocean according to the database maintained by NOAA's Office for Coastal Management. Accurate records of Eastern Pacific tropical cyclones began in 1970. The 2015 version of Tropical Storm Andres formed in a typical location for a May tropical storm.

NOAA predicts an above-average Eastern Pacific hurricane season: 18.5 named storms
Andres is the first salvo in what is likely to be a very busy Northeast Pacific hurricane season. NOAA's pre-season prediction for the Eastern Pacific hurricane season, issued on May 27, calls for 15 - 22 named storms, 7 - 12 hurricanes, 5 - 8 major hurricanes, and an ACE index 110% - 190% of the median. The mid-point of these ranges gives us a forecast for 18.5 named storms, 9.5 hurricanes, and 6.5 major hurricanes, with an ACE index 150% of average. The 1981 - 2010 averages for the Eastern Pacific hurricane season are 15 named storms, 8 hurricanes, and 4 major hurricanes.


Guest forecaster at the NHC: President Obama
Sharp-eyed readers may have noticed an unexpected name at the bottom of this morning's NHC discussion of Tropical Storm Andres. Signing the forecast discussion, along with forecaster Dan Brown, was President Barack Obama. The president visited NHC this morning to receive the annual presidential briefing on the upcoming hurricane season. This was the first time Obama was onsite at NHC for the briefing. Pictured at right: NHC director Rick Knabb, President Obama, NHC/TAFB chief Hugh Cobb, Secretary of Commerce Penny Pritzker, and NOAA Administrator Kathryn Sullivan.

Jeff Masters

Hurricane

Active Atlantic Hurricane Period That Began in 1995 May be Over: NOAA

By: Jeff Masters , 5:14 PM GMT on May 27, 2015

It should be another quiet Atlantic hurricane season in 2015, and the active hurricane pattern that began in 1995 may now be over, said NOAA in their May 27 seasonal hurricane forecast. They give a 70% chance of a below-normal season, a 20% chance of a near-normal season, and only a 10% chance of an above-normal season. They predict a 70% chance that there will be 6 - 11 named storms, 3 - 6 hurricanes, and 0 - 2 major hurricanes, with an Accumulated Cyclone Energy (ACE) 40% - 85% of the median. If we take the midpoint of these numbers, NOAA is calling for 8.5 named storms, 4.5 hurricanes, 1 major hurricane, and an ACE index 62.5% of normal. This is well below the 1981 - 2010 average of 12 named storms, 6 hurricanes, and 3 major hurricanes. Hurricane seasons during the active hurricane period 1995 - 2014 averaged 14.7 named storms, 7.6 hurricanes, and 3.5 major hurricanes, with an ACE index 142% of the median. Only three seasons since 1995 have been classified by NOAA as being below normal--including two El Niño years (1997 and 2009), and the neutral 2013 season.



The forecasters cited the following main factors that will influence the coming season:

1) The current borderline weak/moderate El Niño event is expected to persist or intensify during the 2015 hurricane season. El Niño events tend to suppress Atlantic hurricane activity in three ways:

- By creating high levels of wind shear over the tropical Atlantic, which tends to tear storms apart.
- By increasing sinking motion and high pressure over the tropical Atlantic.
- By making the air more stable over the tropical Atlantic.

2) Near-average sea surface temperatures (SSTs) are in place over the hurricane Main Development Region (MDR), from the Caribbean to the coast of Africa between between 10°N and 20°N. These SSTs are expected to be near or below average during the peak August - October portion of hurricane season, and are expected to be cooler than SSTs in the remainder of the global tropics (SSTs in the remainder of the global tropics were 0.31°C warmer than SSTs in the MDR in May.) This configuration of SSTs is often quite hostile to Atlantic tropical cyclone development.

3) The active period of hurricane activity that began in 1995 due to a natural decades-long cycle in hurricane activity called the Atlantic Multi-decadal Oscillation (AMO) may now be over. The SST pattern associated with that cycle is absent this year, and NOAA said: "There have been two seasons in a row, 2013 and 2014, with below-normal and near-normal activity respectively and neither had an El Niño event responsible for the reduced activity. The current configuration of SSTs in the Atlantic Ocean, both in the MDR and the entire North Atlantic, are suggestive that the AMO may no longer be in the warm phase."


Figure 1. Hurricane Gonzalo as seen from the International Space Station on October 16, 2014. At the time, Gonzalo was at peak strength, with 145 mph winds, and was the first Atlantic hurricane to reach sustained winds of at least 145 mph since Hurricane Igor of 2010. Gonzalo hit Bermuda just a week after Hurricane Fay hit the island, and Gonzalo's remnants went on to batter the United Kingdom on October 21 with wind gusts exceeding 100 mph, killing three people there. Image credit: Alexander Gerst.

Though hurricane activity in the MDR should be limited this year, we often see storms occurring in a belt extending from the Gulf of Mexico to the central Atlantic north of the MDR during El Niño years, in association with reduced vertical wind shear across the region (Goldenberg and Shapiro 1996). NOAA's long-range CFS model predicts the vertical wind shear in this belt to be weaker than average during the August - October peak of hurricane season, which could allow activity to be near the higher end of their predicted ranges (11 named storms, 6 hurricanes, and 2 intense hurricanes.) At present, SSTs are unusually warm in the Gulf of Mexico (1°C - 3°C above average) and in a belt from the Bahamas eastward into the central Atlantic (1 - 2°C above average). Storms that form in these waters are typically weaker than ones that form in the MDR, since they have less time to intensify over warm water.

Standard disclaimer: “A below-normal season doesn’t mean we’re off the hook. As we’ve seen before, below-normal seasons can still produce catastrophic impacts to communities,” said NOAA Administrator Kathryn Sullivan, Ph.D., referring to the 1992 season in which only seven named storms formed, yet the first was Andrew – a Category 5 Major Hurricane that devastated South Florida.


Figure 2. 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 1980 - 2014, 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 (2005 - 2014). 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. Using this metric, NOAA's late May forecasts have about 10% skill over chance, while TSR's have about 20% skill. Image credit: Tropical Storm Risk, Inc (TSR).

TSR predicts a below-average Atlantic hurricane season: 11 named storms
The May 27 forecast for the 2015 Atlantic hurricane season made by British private forecasting firm Tropical Storm Risk, Inc. (TSR) will be out later today, and I will update this post with their latest numbers. Their April 9 forecast called for 11 named storms, 5 hurricanes, 2 intense hurricanes, and an Accumulated Cyclone Energy (ACE) of 56.

CSU predicts a below-average Atlantic hurricane season: 9 named storms
Another quiet Atlantic hurricane season is likely in 2015, said the hurricane forecasting team of Dr. Phil Klotzbach and Dr. Bill Gray of Colorado State University (CSU) in their latest seasonal forecast issued April 9. They called for an Atlantic hurricane season with 7 named storms, 3 hurricanes, 1 intense hurricane, and an Accumulated Cyclone Energy (ACE) of 40. The next CSU forecast is due on Monday, June 1, and will receive a lot of media attention. My April 9 blog post has more on their forecast.

UKMET office predicts a below-average Atlantic hurricane season: 9 named storms
The UKMET office forecast for the 2015 Atlantic hurricane season, issued May 21, calls for below-normal activity, with 8 named storms, 5 hurricanes, and an ACE index of 74 occurring during the period June - November. If we add in Tropical Storm Ana, these numbers come out to 9 named storms and 5 hurricanes. In contrast to the statistical models relied upon by CSU, TSR, and NOAA, the UKMET forecast is done strictly using a dynamical global seasonal model, the Met Office GloSea5 system.

WSI predicts a below-average Atlantic hurricane season: 9 named storms
The May 27 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 a quiet Atlantic hurricane season with 9 named storms, 5 hurricanes, and 1 intense hurricane. WSI cites the expectation of El Niño conditions this fall as a key factor influencing their forecast.

Penn State predicts a below-average Atlantic hurricane season: 8 named storms
The May 11 forecast made using a statistical model by Penn State's Michael Mann and alumnus Michael Kozar is calling for a quiet Atlantic hurricane season with 8 named storms, plus or minus 2.7 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 statistical model assumes that in 2015 the mid-May -0.18°C departure of temperature from average in the MDR will persist throughout hurricane season, the El Niño phase will be moderate, 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 named storms, Actual: 9
2010 prediction: 23 named storms, Actual: 19
2011 prediction: 16 named storms, Actual: 19
2012 prediction: 11 named storms, Actual: 19
2013 prediction: 16 named storms, Actual: 14
2014 prediction: 9 named storms, Actual: 8

NCSU predicts a below-average Atlantic hurricane season: 5 named storms
The April 19 forecast from North Carolina State University (NCSU), is calling for a below-average Atlantic hurricane season with 5 named storms, 2 hurricanes, and 1 intense hurricane.

Coastal Carolina University predicts a below-average Atlantic hurricane season: 8 named storms
The April 17 forecast from Coastal Carolina University is calling for a below-average Atlantic hurricane season with 8 named storms, 4 hurricanes, and 2 intense hurricanes.

Cuban Met service predicts a below-average Atlantic hurricane season: 8 named storms
The May 4 forecast from the Cuban Meteorological Service, INSMET, is calling for a below-average Atlantic hurricane season with 9 named storms and 3 hurricanes. The forecast is based on a statistical prediction model developed by Ballester, González and Pérez (2010).


Figure 3. Departure of temperature from average for May 23 for the Northeast Pacific waters showed large areas of SSTs 1 - 2°C (1.8 - 3.6°F) above average. Image credit: NOAA/NHC.

NOAA predicts an above-average Eastern Pacific hurricane season: 18.5 named storms
NOAA's pre-season prediction for the Eastern Pacific hurricane season, issued on May 27, calls for an active season, with 15 - 22 named storms, 7 - 12 hurricanes, 5 - 8 major hurricanes, and an ACE index 110% - 190% of the median. The mid-point of these ranges gives us a forecast for 18.5 named storms, 9.5 hurricanes, and 6.5 major hurricanes, with an ACE index 150% of average. The 1981 - 2010 averages for the Eastern Pacific hurricane season are 15 named storms, 8 hurricanes, and 4 major hurricanes.

NOAA predicts an above-average Central Pacific hurricane season: 5 - 8 tropical cyclones
NOAA's pre-season prediction for the Central Pacific hurricane season, issued on May 27, calls for an above-average season, with 5 - 8 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.

TSR predicts an above-average Northwest Pacific typhoon season: 11 intense typhoons
The May 6 forecast for the 2015 Northwest Pacific season made by British private forecasting firm Tropical Storm Risk, Inc. (TSR) calls for an active season with 27 named storms, 17 Category 1 or stronger typhoons, 11 intense Category 3 or stronger typhoons, and an Accumulated Cyclone Energy (ACE) of 400. The long-term averages for the past 50 years are 26 named storms, 16 typhoons, 8 intense hurricanes, and an ACE of 294. TSR rates their skill level as modest for these late May forecasts--13% - 28% higher than a "no-skill" forecast made using climatology. TSR’s main predictor for their typhoon season forecast is sea surface temperature (SST) in the region 5 ̊S - 5 ̊N, 140 ̊W - 180 ̊W, which they expect to be 1.0±0.5°C warmer than normal.

Invest 92E likely to become Tropical Storm Andres off the Pacific Mexican coast
The first named storm of the Northeast Pacific hurricane season usually forms by June 10, but we are likely to see Tropical Storm Andres well before that date this year. Satellite loops show that Invest 92E, located about 800 miles south-southwest of Manzanillo, Mexico, has developed a large area of heavy thunderstorms and a good degree of spin. The GFS and European models both show 92E developing into a tropical depression or tropical storm by Saturday. SSTs are a very warm 30°C (86°F), which is nearly 2°C (3.6°F) above average. The 8 am EDT Wednesday SHIPS model forecast predicted that wind shear would remain light over 92E during the coming five days, and I expect 92E will develop into Tropical Storm Andres by early next week. In their 2 pm EDT Wednesday Tropical Weather Outlook, NHC gave 92E 2-day and 5-day development odds of 80% and 90%, respectively. This system is far enough offshore that it will not spread heavy rains to the coast of Mexico over the coming week as it moves west-northwest or northwest, parallel to the coast.

In the Atlantic, the models are depicting high wind shear through June 3 over the majority of the regions we typically see early-season tropical cyclone development--the Caribbean, Gulf of Mexico, and Bahamas. However, late next week we should see a decrease in wind shear, 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 Tropical Storm Bill developing in the Gulf of Mexico or waters near the Bahamas are above average, given the warmer than average SSTs in these waters, and the fact that the Madden-Julian Oscillation (MJO) is expected to be active in the Western Hemisphere during the first ten days of June.

Jeff Masters

Hurricane

Epic Rains, Disastrous Floods Plague Texas, Oklahoma

By: Bob Henson , 2:24 PM GMT on May 26, 2015

Water cascaded through the streets, creeks, and bayous of downtown Austin and Houston on Monday as an upper-level storm inched its way across the southern Great Plains. Slow-moving thunderstorms dumped 6” to 8” across the western Houston metro area between 8:00 and 11:00 p.m., and heavy rains continued well past midnight across much of the south and west metro area, bringing some totals as high as 10+”. Though the Houston flooding came well short of that in 2001’s catastrophic Tropical Storm Allison, countless roads and interstate highways were submerged, and hundreds of homes reportedly took water. This was the latest salvo in a remarkable three-day stretch of torrential rain and destructive flooding across much of Oklahoma and Texas and parts of neighboring states. As of Tuesday morning, the floods had taken at least 8 lives, with at least 12 people missing, and damaged or destroyed many hundreds of buildings.


Figure 1. Damage along River Road in Wimberley, Texas, from a devastating flash flood on Saturday night, May 23. Photo credit: Jerry Lara/The San Antonio Express-News, via AP.

Some of the worst damage occurred in and near the much-touristed town of Wimberley, located along the Blanco River in the Texas Hill Country west of Austin. After a mesoscale convective system (MCS) dropped upward of 7” upstream of Wimberley on Saturday afternoon, the river surged to a new record high in spectacular fashion later that night (see Figure 2 below). The Hill Country has a tragic history of flash flooding, and prompt evacuations no doubt saved many lives. This weekend also saw several “flash flood emergency” declarations by the National Weather Service. This wording is reserved for the highest-end events, much like the “tornado emergency” tag issued when substantial numbers of people are in particular danger. From Saturday through Monday, flash flood emergencies were declared in parts of the Tulsa, Austin, and Houston areas, as well as five counties in west-central Oklahoma. It was the first time that Tulsa and Houston had ever been placed under such alerts.



Figure 2. The level of the Blanco River at Wimberly soared from a more 5 feet on Saturday afternoon, May 23, to a record 40.21 feet at 1:01 am CDT Sunday, more than 7 feet above the prior record. The river gauge stopped reporting after this point. Image credit: NWS Advanced Hydrologic Prediction Service.


Figure 3. Infrared satellite imagery shows the extremely cold cloud tops (gray) associated with the torrential rains in the Houston area at 9:15 pm CDT on Monday night, May 25. Cloud-top temperatures at the height of the storms were colder than -100°F. Note the two eerie faces visible across southeast Texas! Larger version here [with time code but without state boundaries]. Image credit: NASA Earth Science Office, courtesy Stu Ostro.


The scope and intensity of this weekend’s rains resulted from a rich feed of low-level moisture pumping into the region from the Gulf of Mexico, combined with the excruciatingly slow movement of the large-scale storm system. Upper-level winds were largely aligned with the low-level frontal zone, an ideal setup for “training echoes” (successive downpours over the same area). In addition, rains were concentrated by several mesoscale convective vortices, small-scale centers of low-pressure that developed along the frontal bands, in some cases resembling mini-tropical cyclones.


Figure 4. Damaged homes stand next to others that were razed when a powerful tornado tore through a seven-block area in Ciudad Acuna, Mexico, on Monday morning, May 25. Image credit: AP.

A deadly tornado strikes northern Mexico at daybreak
Although flooding was the main threat across Texas and Oklahoma, nearly 100 tornado reports were logged nationwide by the NOAA Storm Prediction Center over the three-day holiday weekend. The deadliest twister so far this year in North America struck on Monday morning in Cuidad Acuña, Mexico, a city of about 200,000 located on the Rio Grande across from Del Rio, Texas. With reported wind estimates in the EF4 range, the tornado killed at least 13 people and injured at least 150, with more than 1,000 homes damaged. [Update: Maria de Jesus Guadalulpe Gallo Banda, from Mexico's National Water Commission, has estimated that peak gusts were 220 kilometers per hour (137 mph), which would correspond to an EF3 rating, or an F2 rating in the original Fujita scale, which is still used in Mexico.] The powerful twister emerged around 7:00 am CDT from an isolated supercell thunderstorm that likely benefited from very strong dynamics associated with an upper-level jet streak plowing across Mexico (see Figure 4). Monday’s disaster brings to mind a similar event that occurred 60 miles to the southeast of Cuidad Acuña on April 24, 2007, when a tornado produced F4 damage in Piedras Negras, Mexico, killing 3 people, and EF3 damage across the Rio Grande in Eagle Pass, Texas, where 7 people died. In general, North American tornadoes this strong are very unusual at such a low latitude. No F5/EF5 events have ever been confirmed this far south in the U.S. or Mexico, and TornadoHistoryProject.com shows that only 10 of 555 U.S. tornadoes rated F4 or EF4 between 1950 and 2014 occurred near or south of 30°N. Six of these were in Texas, two in Louisiana, and two in Florida. Half of them occurred during El Niño conditions (about twice the number you would expect based on El Niño frequency alone), and half occurred between 1:00 am and 11:00 am local time. The strong subtropical jet streams associated with El Niño typically bring very cold upper-level air, which enhances instability, and the powerful jet-level winds help ventilate the rising air within thunderstorms. Moreover, at these low latitudes, surface temperature and moisture values can be high enough to produce very unstable air even during the late night and early morning.


Figure 5. WunderMap analysis of flow at the 200-hPa level (about 39,000 feet) at 1200 GMT on Monday, May 25. A powerful subtropical jet can be seen flowing eastward across Mexico, then northward through the Mississippi Valley. The tornadic supercell at Cuidad Acuna, Mexico, occurred about an hour before the time of this analysis, near the left front corner of a powerful “jet max” (the greenish strip across Mexico). Severe storms tend to occur most often at the left front and right rear of jet maxima, where upward motion is favored. Also evident is the mammoth ridge extending from western Canada into the Arctic Ocean, where record-high temperatures have been observed in recent days. The northwestern ridge and western U.S. trough have persisted through much of May.

Precipitation persistence: the story of 2015
One of the most intriguing questions in climate change research is whether blocking-type patterns might be fostered by rising global temperatures and the resulting effects on jet-stream behavior. Over the last month, the same pattern favoring heavy rain across Texas and Oklahoma has kept rain away from the Northeast. It was just three months ago that an unprecedented month-long stretch of heavy snowfall brought Boston and much of New England to its frosty knees. A couple of individual snowstorms within that stretch were among Boston’s heaviest, but it was the relentlessness of the cold, snowy conditions that truly stood out and caused such misery. Likewise, the unrelenting rainfall across the southern Plains this month has caused pile-on effects, as downpours flow off saturated soil and farmers struggle to get spring crops planted. During the 30 days ending on May 25, Norman, OK, received an astounding 24.10”. Oklahoma is now assured of its wettest month on record, according to the Oklahoma Climatological Survey.


Figure 6. 30-day precipitation from 1200 GMT April 25 through May 25, 2015. Some parts of southern Oklahoma and far north Texas have received more than 20” of rain, while central New England has recorded less than half an inch. Image credit: NWS Advanced Hydrologic Prediction Service.

Here are some individual locations that have already set records for their wettest May as of midnight Monday night, with nearly a week left to go in the month. (Thanks to Nick Wiltgen at the Weather Channel for compiling these statistics.)

Oklahoma City, OK: 18.85” (previous May record 14.52” in 2013; previous all-time record 14.66” in June 1989)
Fort Smith, AR: 18.07” (previous May record 13.67” in 1943; previous all-time record 15.02” in June 1945)
Austin, TX (Camp Mabry): 16.72” (previous May record 14.10” in 1895; all-time record 20.78” in Sept. 1921)
Wichita Falls, TX: 14.15” (previous all-time record 13.22” in May 1982)

Meanwhile, back in the Northeast, the tables have turned in a way that might have seemed hard to fathom three months ago. The most recent U.S. Drought Monitor shows moderate drought (D1) covering all of Massachusetts, Connecticut, and Rhode Island, as well as parts of New York, Maine, and Pennsylvania. Even the enormous snowfalls of February didn’t contain especially large amounts of moisture compared to what can fall in a midwinter New England rainstorm, and the tap hasn’t been flowing much at all lately. In the 30 days ending on May 26, less than an inch of rain fell across most of New Jersey, southeast New York, and southern New England, with some locations getting little more than sprinkles. Further south, only a trace has fallen this month in Charlotte, North Carolina. Reservoirs and water tables in New England are still in relatively good shape, but streamflows are quite low. Potential records for driest or second-driest May at locations with more than a century of data include:

Concord, NH: 0.07” (record low 0.50”, 2008; records begin in 1903)
Providence, RI: 0.51” (record low 0.57”, 1939; records begin in 1904)
Hartford, CT: 0.60” (record low 0.73”, 1959; records begin in 1905)
Boston, MA: 0.31” (record low 0.25”, 1944; runner-up 0.32”, 1903; records begin in 1872)
Albany, NY: 0.31” (record low 0.15”, 1903; runner-up 0.73”, 1920; records begin in 1874)
New York, NY (Central Park): 0.32” (record low 0.30”, 1903; runner-up 0.34”, 1887; records begin in 1871).

A change in the prevailing upper-level features toward more zonal (west-to-east flow) across North America, as outlined by Steve Gregory in his latest blog, may help put some of these dry-May records out of reach. A midweek front will bring scattered showers and thunderstorms, with a better chance of widespread wetting rains over the weekend into early next week as a front stalls across the mid-Atlantic.

Watch for a report from Jeff Masters on NOAA’s Atlantic hurricane outlook for 2015, scheduled for release on Wednesday. We will also be covering the devastating pre-monsoonal heat wave that has taken hundreds of lives across India.

Bob Henson


Figure 7. Departures from average outgoing longwave radiation (OLR) for the period from April 23 to May 22, 2015. OLR is maximized when skies are clear and minimized when outgoing radiation is blocked by clouds and precipitation. A belt of lower-than-average OLR values extends from the eastern tropical Pacific (where El Niño has warmed the sea surface) to the central and western United States, illustrating the likely connection between El Niño and this month’s heavy rainfall. See the AGU blog by Dan Satterfield for more on this apparent link. Image credit: NOAA/ESRL/PSD, courtesy Eric Blake.

Additional footage at Barton Springs Rapids, umm... Pool.

Posted by Austin Parks and Recreation Department on Sunday, May 24, 2015

Video 1. Barton Springs Rapids, on Barton Creek within the city limits of Austin, lived up to its name on Sunday, May 24, prior to the area’s second round of major flooding on Monday. If embedded video does not play, it can be viewed here. Image credit: Austin Parks and Recreation Department.


Figure 8. An ironic scene from the town of Cache, in southwest OK, on Sunday, May 24. Image credit: wunderphotographer lmalcolm.

Flood Tornado Drought

Summer Weather Watch: Keep an Eye on These Five Possibilities

By: Bob Henson , 4:57 PM GMT on May 22, 2015

It’s Memorial Day weekend, the traditional start of the U.S. summer season, and millions are wondering what kind of weather the next three months will bring. Seasonal predictions have their limits any time of year, and that’s especially true in summer, when upper-level winds are weaker and local influences play a larger role. Moreover, the largest single influence on year-to-year climate variability--the El Niño/Southern Oscillation (ENSO)--is often at low ebb in the northern summer. Not so this year. An unseasonable El Niño event is now approaching moderate strength and is projected to continue intensifying through the summer, perhaps reaching record or near-record strength for the time of year by August. Instead of the typical lack of a summertime push from El Niño or La Niña, we’re thus left with a much different kind of prediction challenge: a summer setup so unusual that we have few analogs to go by. With that caveat, I’ll stick my neck out and offer a Top Five List (with apologies to David Letterman) of things I’ll be watching for as the lazy, crazy, and occasionally hazy days of summer unfold.



Figure 1. Departures from average temperature across the U.S. for the summers of 1982 and 1997, both of which led into strong El Niño events. Image credit: NOAA/ESRL/PSD.


Figure 2. The weather prediction firm WSI is calling for relatively mild conditions this summer across the bulk of the United States, with unusual heat confined mainly to the western U.S. and Florida. Image credit: WSI.


Cool, man, cool
The summers of 1982 and 1997, which preceded the two strongest El Niño events on record, were cooler than average across most of the United States (see Figure 1 above). No analog is perfect, but based in part on the patterns observed in those two years, “we expect the weakest nationwide cooling demand since at least 2009,” says WSI in its summer energy outlook for 2015 (Figure 2). Other years with at least a moderate Oceanic Niño Index value (at least +1.0) in Jun-Jul-Aug include 1972, 1965, and 1957; all but 1957 had widespread below-average summer temperatures. Precipitation signals for the summer are less straightforward, although during winter El Niño tends to bring wetter-than-average conditions across the southern half of the United States. The strong subtropical jet stream that’s fed much of the low-latitude U.S. rainfall over the last month may weaken as we get into summer, then restrengthen in the fall, but signals remain positive for widespread summer moisture. The average of a variety of climate models assembled through the North American Multi-Model Ensemble (NMME) suggests relatively wet conditions across much of the nation, with a cooler-than-average pocket in the nation’s heartland and relatively warm temperatures close to the coasts. These tendencies are reflected in the National Weather Service summer outlook (see Figure 3 below). The central U.S. already has a head start toward a fairly mild summer due to the extremely wet conditions across most of the Plains over the last month. Even when the rains abate and the summer sun kicks in, some of that energy will go toward evaporating surface-based moisture, rather than heating up the ground and the surface air.




Figure 3. Seasonal predictions from the National Weather Service (June-August) showing where the odds are leaning for temperature (left) and precipitation (right). “EC” denotes equal chances of above- or below-average conditions. Image credit: NWS/Weather Prediction Center.


For fire and heat, head northwest
Landscapes are parched from most of California up to the interior of Alaska, as well as adjacent northwest Canada. A major high-latitude heat wave sent temperatures on Thursday in Barrow, AK, up to 47°F, the warmest ever observed so early in the season and only the second time that temperature has been reached before June (more here from the Weather Channel’s Jon Erdman). The warm temperatures have triggered unprecedented flooding that’s closed more than 50 miles of the Dalton Highway, a key route through northern Alaska. It may be a particular rough season for wildfires across those higher-elevation forests where snow was extremely scant this past winter, from the Sierras north through the Cascades and into British Columbia.

Hurricanes aplenty in the northeast Pacific
All signs point toward a blockbuster year for tropical cyclone activity in the northeast Pacific basin, a region favored by the proximity to unusually warm sea-surface temperatures (SSTs) associated with El Niño. SSTs are already more than 1°C (1.8°F) above average over a large swath of the region where northeast Pacific hurricanes typically form. This morning the National Hurricane Center is investigating two potential tropical cyclones for possible development in the northeast Pacific, and models suggest that one of these could become the the farthest-west tropical cyclone on record for this basin so early in the year. It’s not out of the question that at least one northeast Pacific tropical cyclone will move toward Baja California later this summer, perhaps bringing heavy rain and wind to the southwest U.S. In addition, the northwest Pacific is off to a rip-roaring season, with three Category 5 storms to date, the most ever recorded so early in the year. Conditions should lean toward more tranquility across the North Atlantic, where El Niño effects will likely lead to a below-average year for tropical cyclone activity. NOAA will issue its Atlantic hurricane outlook on Wednesday, May 27.


Figure 4. Unusually early warmth now across Alaska and western Canada may push into the Arctic next week, as shown by temperature anomalies predicted by the 0000 GMT Thursday run of the GFS model for 0000 GMT Wednesday, May 27. Image credit: ClimateReanalyzer.com/University of Maine.


Another crucial year for Arctic sea ice
The extent of sea ice in and near the Arctic is already close to record-low values for this time of year. Now that we’re approaching the summer solstice, a key variable that will shape ice melt is the extent of cloudiness over the next few weeks; clear skies allow the round-the-clock sunlight of the midsummer Arctic to have maximum impact on the ice. Long-range models project that the heat wave across northwest Canada and Alaska will push still further north next week, sending temperatures close to or above the freezing mark across a large swath of the western Arctic weeks ahead of schedule. This could result in widespread formation of melt ponds atop the ice, which absorb additional heat from the sun and hasten further melting. It’s too soon to say for sure--the weather in June and July will play a huge role--but I’d consider a new record-low extent for Arctic ice a real possibility for 2015.

Taking aim at a new global heat record
This year is already the warmest in global records for the period January through April. El Niño tends to warm Earth’s atmosphere (by suppressing the upwelling of cold water and spreading oceanic heat over a large area, where it can warm the air above). So barring any major volcanic eruptions, we can expect the next three months to keep this year rolling toward a likely, if dubious, victory over 2014 in the troublesome global-heat race.

Across the soaked Plains, a memorable Memorial Day weekend
Just when it seemed more water couldn’t possibly fall over Texas and Oklahoma, yet another Pacific storm system will wring out more heavy rains this weekend. Flash flood watches are already posted for much of the two states, plus western Arkansas, with five-day rainfall amounts likely to top 3” over large areas and exceed 6” in spots. With the rains juxtaposed over some locations that have received 10” to 20” over the last several weeks, there’s a high potential for dangerous flash flooding this weekend, as well as major river flooding in the days that follow. Severe weather should be relatively subdued this weekend, with a few tornadic storms possible in scattered pockets as the upper low now over the Southwest slowly progresses east. Outside of the plains, travel impacts should be relatively minimal, although showery conditions in many areas will dampen more than a few outdoor activities. Temperatures will be a bit cooler than average over much of the West, while an early-season “warm wave” begins building toward the East Coast.

Have a great holiday weekend, everyone!

Bob Henson



Figure 5. A thunderstorm over central Pennsylvania on Sunday, May 17. Image credit: wunderphotographer baxtheweatherman.


Long-Range Summer Outlook

What Can Lightning Tell Us about Storm Severity

?

By: Bob Henson , 5:42 PM GMT on May 21, 2015

As soon as an incipient thunderstorm spits out its first cloud-to-ground lightning bolt, it’s a potentially deadly threat. A complex of severe storms can generate many thousands of lightning flashes per hour. Along with being a hazard in its own right, lightning can serve as an useful clue to how quickly a thunderstorm is strengthening. New tools along these lines have been developed to help forecasters, and they’re being tested this spring at NOAA's Hazardous Weather Testbed at the National Weather Center in Norman, OK, where I visited earlier this month.




Figure 1. Lightning and the setting sun paint a colorful scene over Washington, D.C., on April 20, 2015. Image credit: wunderphotographer wolfpackwx.

Those of you in warm climates may have seen a weak shower suddenly deposit a burst of rain just minutes after you heard the first clap of thunder. For decades, this anecdotal evidence couldn’t be reliably examined. The advent of lightning detection systems in the 1980s meant that lightning flashes could be quantified for the first time. Most of the country is now covered by a network of sensors that detect the location and timing of cloud-to-ground lightning (CGs) by measuring the electromagnetic signals released at ground level. There are also regional lightning mapping arrays (LMAs) that detect intracloud lightning (ICs) by sensing the very high frequency radiation released at each “elbow” of a zigzagging flash. Together, these systems allow what's called total lightning (CGs + ICs) to be quantified. Some of the earliest work using these data found intriguing connections to severe weather. A 1989 study led by Don MacGorman (National Severe Storms Laboratory) found that the total lightning flash rate in one Oklahoma thunderstorm peaked about 5 minutes before a tornado developed. In another early study, Steven Goodman (NOAA/NESDIS) and colleagues discovered a spike in the total flash rate of an Alabama thunderstorm that preceded small hail and a microburst by several minutes. Goodman and Patrick Gatlin (University of Alabama in Huntsville) also developed the first algorithm designed to automatically identify lightning jumps.

More recently, Christopher Schultz (now at NASA) led a series of studies looking more broadly at whether lightning-jump algorithms could serve as an early-warning tool to help forecasters identify which storms might soon become severe. The results were highly encouraging, especially when the algorithm flagged jumps in total lightning that were two standard deviations greater than the rate observed a few minutes earlier. In a 2011 study of more than 700 thunderstorms (mostly in northern Alabama), this “two-sigma” index detected 79% of all severe thunderstorms, with a relatively low false-alarm rate of 36%. Schultz’s recent dissertation work demonstrated that lightning jumps, as opposed to more general increases in flash rate, were closely related to increases in the storm’s peak updraft speed, and in the storm volume that features updrafts classified as intense (e.g., at least 10 meters per second or 22 mph). These increases in updraft size and speed preceded lightning jumps by roughly 4 to 12 minutes.

Lightning jumps can’t tell us exactly what kind of severe weather a storm will produce, and not every lightning jump will lead to a severe outcome. Moreover, some thunderstorms reach severe levels without producing huge amounts of lightning. However, in general, lightning jumps can provide an almost-instantaneous measure of how quickly a thunderstorm updraft is strengthening. As the vertical motion in a storm intensifies, ice crystals, supercooled water droplets, and graupel (snow pellets)--together referred to as mixed-phase precipitation--grow more rapidly and bump into each other more readily, transferring charge among each other. Different speeds of descent allow for the charged particles to assemble in zones of positive and negative charge, which increases the storm’s ability to generate lightning. In a newly forming thunderstorm, the strengthening updraft typically produce the charge separation needed for a lightning jump a few minutes before it has time to generate one or more of the markers that the National Weather Service uses to classify a thunderstorm as severe: hail larger than 1” in diameter, winds reaching 58 mph, and/or a tornado. (Interestingly, lightning itself is not one of the elements that officially define a storm as severe, in part because it’s traditionally been so hard to quantify.) “The lightning jump provides forecasters vital information on the growth of the mixed-phase updraft size and speed within the thunderstorm, which is one crucial component that forecasters seek during the warning decision-making process,” says Schultz.


Figure 2. The solid dark-orange blob indicates a four- to five-sigma jump in lightning activity at 2009 GMT on Monday, May 18, in Tillman County, OK. The storm grew quickly, and several tornadoes were reported in Wichita County, TX, and Tillman County, OK, between 2030 and 2100 GMT. Image credit: GOES-R Proving Ground Blog.


Satellite-based measurements of total lightning are a promising adjunct to ground-based networks, which leads us to this year’s spring experiment in Oklahoma. The project is testing a variety of storm-intensity clues that will soon be provided by the GOES-R series of satellites, which are scheduled for launch beginning in the spring of 2016. GOES-R is set to include a Geostationary Lightning Mapper (GLM) that will use optical sensing to quantify total lightning at a spatial resolution of around 10 km, with a mere 20 seconds needed for data processing. In place of these yet-to-be-launched instruments, ground-based mapping array data from Earth Networks are being used in this spring’s experiment as proxies to replicate what GOES-R will be able to provide. Using this "pseudo-GLM," or PGLM, researchers are testing a lightning jump algorithm that notifies forecasters when a storm is showing jumps at various sigma levels, based on lightning rates across 1- and 6-minute intervals. The forecaster can then keep a closer eye on a particular storm if the lightning jump is dramatic enough, and if it's quickly backed up by other evidence, such as satellite imagery or radar data. “It’ll help forecasters in terms of situational awareness--what storms they should focus on,” says Geoffrey Stano (NASA/ENSCO), who's been helping to coordinate the spring experiment.

Even a few minutes of extra notice may be important with a fast-growing storm, especially if it has the potential to cause serious trouble. In a case study of the storm that produced a deadly tornado in Newcastle and Moore, OK, on May 20, 2013, Stano and colleagues found that one distinct lightning jump preceded the formation of severe hail by 19 minutes. A second jump occurred 26 minutes before the tornado formed, indicating that the storm’s updraft speed and size had increased again and that the potential for severe weather was still present. Lightning jumps may also be helpful in identifying the few storms that have severe potential on a day that's otherwise marginal.

Many examples of how forecasters are using the lightning-jump algorithm, and other tools being tested this spring, can be found at the GOES-R Proving Ground Blog at the Hazardous Weather Testbed.

This week's WunderPoster: Shazam!
Lightning strikes again in today's blog post, this time as the star of the latest in our WunderPoster series. Created by WU’s Skyler Rexrode, this depiction of lightning's power and majesty was inspired by a lighting flash in Wyoming photographed by Bryan Downie (@b_down13). Today's installment concludes our initial set of WunderPosters, but all 16 of the posters produced to date will continue to be available for downloading in formats suitable for posters or postcards. Our thanks go to the hundreds of community members who submitted images for the community-inspired WunderPosters that debuted this month. Hats off to the entire WU design team as well!

We'll be back on Friday with coverage of the weather we can expect this holiday weekend and what the upcoming summer may bring us.

Bob Henson


Figure 3. A severe thunderstorm over California’s Antelope Valley is laced with intracloud lightning. Image credit: wunderphotographer brandyn.

Severe Weather Lightning

Earth has its 4th Warmest April and Warmest January - April Period on Record

By: Jeff Masters , 3:35 PM GMT on May 19, 2015

April 2015 was Earth's fourth warmest April since global record keeping began in 1880, said NOAA's National Climatic Data Center (NCDC) on Tuesday. NASA rated April 2015 as the 3rd warmest April on record. April 2015's warmth makes the year-to-date period (January - April) the warmest such period on record, and the past twelve months the warmest 12-month period in recorded history. Nine of the ten warmest 12-month periods have occurred within the past two years (the exception: September 1997–August 1998 was tied for eighth warmest.) Global ocean temperatures during April 2015 were the warmest on record, and global land temperatures were the 10th warmest on record. Global satellite-measured temperatures in April 2015 for the lowest 8 km of the atmosphere were the 9th or 16th warmest in the 37-year record, according to the University of Alabama Huntsville (UAH) and Remote Sensing Systems, respectively.


Figure 1. Departure of temperature from average for April 2015, the 4th warmest April for the globe since record keeping began in 1880. A few areas saw record warmth, including Cuba, part of southern Mexico, small parts of west central South America and southern Brazil, regions of Mauritania, and part of eastern Tanzania. Part of north central Russia had April temperatures more than 5°C (9°F) above average. Cooler-than-average temperatures were notable across large portions of central to western Australia, parts of South Asia, central North Africa, and part of eastern Canada. Record cold was not observed over any land areas. Image credit: National Climatic Data Center (NCDC) .


Figure 2. Still frame from a YouTube video by e2theloc of the EF4 tornado that hit Rochelle, Illinois on April 9, 2015. The tornado was part of a four-day severe weather outbreak that killed 3 people and did over $1 billion in damage. This has been the only EF4 tornado in the U.S. thus far in 2015.

One billion-dollar weather disaster in April 2015: Midwest U.S. severe weather April 7 - 10
One billion-dollar weather-related disaster hit the Earth last month, according to the April 2015 Catastrophe Report from insurance broker Aon Benfield: a severe weather outbreak on April 7 - 10 in the U.S. Plains, Midwest, and Mississippi Valley that caused at least $1 billion in damage. So far in 2015, there have been just three billion-dollar weather disasters. However, a fourth--the intensifying drought in the Western U.S.--is likely to cause $3 billion in agricultural damage in California, and another $1.2 billion in damage in the State of Washington. The deadliest weather-related disaster of April 2015 was a massive landslide that hit Badakshan village in northern Afghanistan on April 27, killing 52 people and destroying almost 100 homes. The slide was triggered by a big snowfall event.



Arctic sea ice falls to 2nd lowest April extent on record
Arctic sea ice extent during April 2015 was the 2nd lowest in the 36-year satellite record, according to the National Snow and Ice Data Center (NSIDC). April was marked by surface air temperatures about 1 - 3°C (2 - 5°F) above average throughout the Arctic, except for Greenland and the Canadian Archipelago, where temperatures were 1 - 3°C (2 - 5°F) below average.

Notable global heat and cold marks set for April 2015
Hottest temperature in the Northern Hemisphere: 48.0°C (118.4°F) at Larkana, Pakistan, April 20
Coldest temperature in the Northern Hemisphere: -55.4°C (-67.7°F) at Summit GEO, Greenland, April 11
Hottest temperature in the Southern Hemisphere: 41.5°C (106.7°F) at Borroloola, Australia, April 6
Coldest temperature in the Southern Hemisphere: -76.6°C (-105.9°F) at Concordia, Antarctica, April 22

Major stations that set (not tied) new all-time heat or cold records in April 2015
Cocos Islands (Australia), max. 32.8°C, April 8 (Tied all-time territorial record high)
Navrongo (Ghana), max. 43.3°C, April 10 (New all-time national record high for Ghana)
Ba Don (Vietnam), max. 40.5°C, April 19
Thakhek (Laos), max. 42.0°C, April 20 (Tied all-time national record high for Laos)
Nakhon Phanom (Thailand), max. 42.1°C,  April 20
Roi Et (Thailand), max. 42.2°C, April 21
Nong Phlub (Thailand), max. 41.8°C, April 21
Havana (Cuba), max. 37.0°C, April 26
Holguin (Cuba), max. 38.7°C, April 26
Guaro (Cuba), max. 38.0°C, April 26
Merida (Mexico), max. 43.6°C, April 26
Contramaestre (Cuba), max. 37.7°C, April 27
Velasco (Cuba), max. 38.6°C, April 28
Ciego de Avila (Cuba), max. 38.0°C, April 28
Tela (Honduras), max. 40.6°C, April 28
Coro (Venezuela), max. 43.6°C, 29 April 29 (New all-time national record high for Venezuela)
Puerto Padre (Cuba), max. 38.4°C, April 29
Punta Lucrecia (Cuba), max. 37.3°C, April 29
Nuevitas (Cuba), max. 38.5°C, April 30



New all-time national and territorial heat records set or tied in 2015
As of May 18, 2015, eight nations or territories have tied or set all-time records for their hottest temperature in recorded history thus far in 2015, and one (Israel) has set an all-time cold temperature record. For comparison, only two nations or territories set all-time heat records in 2014, and nine did in 2013. The most all-time national heat records in a year was nineteen in 2010 (21 records at the time, but two have been broken since.) Most nations do not maintain official databases of extreme temperature records, so the national temperature records reported here are in many cases not official. I use as my source for international weather records researcher Maximiliano Herrera, one of the world's top climatologists, who maintains a comprehensive list of extreme temperature records for every nation in the world on his website. If you reproduce this list of extremes, please cite Maximiliano Herrera as the primary source of the weather records. Wunderground's weather historian Christopher C. Burt maintains a database of these national heat and cold records for 235 nations and territories on wunderground.com's extremes page. Here are the all-time national heat and cold records set so far in 2015:

Palau tied its national heat record of 34.4°C (94.0°F) at Koror Airport on May 14.
Venezuela set a new national heat record of 43.6°C (109.9°F) at Coro on April 29. Previous record: 42.0°C (107.6°F) at Machiques, Zulia State in February 1983.
Laos tied its national heat record of 42.0°C (107.6°F) at Thakhek on April 20.
Ghana set a new national heat record of 43.3°C (109.9°F) at Navrongo on April 10. This is the third time this year Ghana has tied or set a new all-time heat record. Previous records: 43.1°C (109.6°F), set the previous day, on April 9, and 43.0°C (109.4°F) on February 12.
Cocos Islands (Australian territory) tied their all-time heat record with 32.8°C (91.0°F) on April 8.
Antarctica set a new territorial heat record of 17.5°C (63.5°F) at Esperanza Base on March 24. Previous record: 17.4°C (63.3°F) at Marambio Base, set the previous day. The World Meteorological Organization (WMO) has appointed a committee to study this event and determine if this represents an official record for the continent.
Equatorial Guinea set a new national heat record of 35.5°C (95.9°F) at Bata on March 17. Previous record: 35.3°C (95.5°F) at Malabo in February 1957.
Wallis and Futuna Territory (France) set a new territorial heat record with 35.5°C (95.9°F) on January 19 at Futuna Airport.

Israel set a new national cold record of -14.2°C (6.4°F) at Merom Golan on January 10.

Mr. Herrera originally listed Samoa as tying its national heat record with 36.5°C (97.7°F) on January 20 at Asau, but a subsequent review of the record revealed possible issues with the measurement equipment, so this record is dubious.


Figure 3. This image of the extent of the Saharan Air Layer (SAL) created at 2 pm EDT Wednesday, April 29, 2015, shows that dry, dusty air flowing eastwards from the Sahara Desert was over the Caribbean and northern South America. The associated dry conditions over Venezuela helped bring the hottest temperature ever recorded in that country, 43.6°C (109.9°F) at Coro. Image credit: University of Wisconsin/CIMSS.

Jeff Masters

Climate Summaries

Drenchings Trump Twisters in Tornado Alley

By: Bob Henson , 8:42 PM GMT on May 18, 2015

Despite a few high-profile tornadoes on Saturday, damage and injuries were minimal, and millions of people across the nation’s heartland experienced a more garden-variety weekend of spring showers and thunderstorms, some bearing extremely heavy rain. The unusually wet pattern will continue over the Plains this week, with severe weather mostly restricted to Texas until at least Friday. The most impressive supercell storms on Saturday were across northwest Texas and southwest Oklahoma, with the most prominent tornado traveling alnog a 35-mile-long path from near Elmer to near Snyder. Hitting mostly rural areas, this large tornado earned a preliminary rating of EF2 on the Enhanced Fujita Tornado Damage Scale, based on initial surveys coordinated by the National Weather Service office in Norman, OK. It’s possible that considerably stronger winds existed but failed to strike any of the objects (“damage indicators”) that are used in storm surveys to derive an EF rating. Storm-chaser video from near Elmer showed rapid motion around a wedge-shaped tornado that at one point appeared several times wider than it was tall (see Video 1 below). This is largely a function of the very low cloud base produced by high relative humidity near the ground.


Figure 1. A panoramic view of the storm in southwest Oklahoma that generated the tornado shown in Video 1 and Figure 2. Image credit: Daphne LaDue.


Video 1. Impressive footage from the Elmer-Tipton tornado. The dramatic width of the tornado during its “wedge phase” is clearly evident. If embedded clip is not visible, the video can be viewed directly at YouTube. Video credit: Jeff Snyder.


Figure 2. The “stovepipe phase” of the long-lived tornado in southwest Oklahoma on Saturday, May 16. This photo was taken near the town of Tipton. Image credit: James LaDue.


The University of Oklahoma’s RaXpol radar, a mobile, high-resolution dual-polarization system, gathered data from the Elmer-Snyder tornado. Although the radar was located roughly 10 to 12 miles from the tornado, the data will help clarify wind speeds just above the surface in and near the twister. (Note that radar-derived winds are averaged over a three-dimensional volume around the beam, rather than derived at a single point.) From an initial scan of the data, Howie Bluestein (University of Oklahoma) estimates that peak winds from RaXpol were around 150 mph in the lowest few hundred feet, with evidence of debris in the dual-polarization data at altitudes at least as high as 25,000 feet. “There was a classic debris signature aloft,” says Bluestein. Researchers are pondering ways to rate tornadoes that could employ the growing availability of high-resolution radar estimates of tornado winds without compromising the consistency of the nation’s tornado database, in which most events cannot be linked to such radar data. The El Reno tornado of May 31, 2013, was officially rated EF3, based on the available damage indicators, while RaXpol detected EF5-strength winds of at at least 290 mph over a volume centered less than 60 feet above ground level. (See this Weather and Forecasting journal article for more details on the 2013 RaXpol data.)

Oklahoma was on high alert Saturday night after the highly visible tornadoes appeared on local and national TV. Fortunately, the initial supercells congealed into a solid line before reaching central Oklahoma, which reduced their ability to form significant tornadoes, but a new broken line of supercells formed a bit further east. A late-night tornado (11:30 pm CDT) produced EF2 damage along a path at least 12 miles long near the Tulsa suburb of Broken Arrow. In all, Saturday produced 45 tornado reports across 9 states, with about half occurring in Texas and Oklahoma. As was the case on the previous Saturday (May 8), the day’s severe weather was split into two focal points by a large area of morning showers and thunderstorms that developed early in the day over northern Oklahoma and Kansas, keeping the atmosphere in those areas too cool and stable for supercell activity. The other hot spot on Saturday was in central Minnesota, where 9 tornadoes were reported, including one that struck a farmstead near Montevideo.


Figure 3. The 0- and 84-hour forecasts at 500 mb from the NAM model, valid at 1200 GMT Monday, May 18, and 0000 GMT Friday, May 22, are remarkably similar, reflecting the persistence in the current pattern. Image credit: NCAR/RAL Real-Time Weather Data.


Another wet week ahead
Prodigious rains are emerging as the big weather story this month—not a surprise, given that an unseasonably early El Niño event is now strengthening. Persistent, recurrent upper lows in the Southwest (see Figure 3) have channeled deep moisture from the Gulf of Mexico across the central states and provided jet-stream energy to lift the moist air. It appears quite possible that this month will eclipse June 1989 as the wettest month in Oklahoma City history, in records going back to late 1890. With 12.85” reported as of Monday morning, only 1.82” is needed to break the 1989 record, with more rain in the forecast throughout this week. Other cities are also in the running for monthly rain records, including San Diego, where 2.35” has fallen this month; the city’s May record is 2.54” (1921), and light showers are expected late this week. Phoenix scored 0.93” of rain last Thursday, the most observed on any May day on record. The current monthly total of 1.17” in Phoenix makes it the second-wettest May on record, behind only May 1930 (1.31"). A wet weekend pushed Jamestown, ND, up to 8.29” for the month, the highest May total in records going back to 1893.


Figure 4. The 7-day precipitation outlook from NOAA’s Weather Prediction Center for the period 1200 GMT Monday, May 18 - 25. Image credit: NOAA/NCEP Weather Prediction Center.

The zone of rich moisture that’s covered the Plains much of this month will be pushed southward early this week by a chilly-for-May surface air mass. The next chance for widespread severe weather will likely be on Friday or Saturday, as yet another upper low in the Southwest (see Figure 3, above) moves across the Plains. However, it’s again possible that early-morning showers and storms will shunt the most unstable air away from the zone of strongest wind shear, limiting the area at risk of significant tornadoes or other severe weather. Heavy rains appear to be a safe bet (see Figure 4, above), again putting parts of the Southern Plains at risk of flash flooding.

Bob Henson


Figure 5. A “ground-to-cloud” flash lights up the sky near Bixby, OK, on Saturday, May 16. Upward-directed lightning, which typically occurs near transmission towers, wind turbines, and other large objects, has recently been linked to thundersnow events. Image credit: wunderphotographer mrwing13.

Tornado

Super Typhoon Dolphin Becomes Earth's 5th Category 5 Storm of 2015

By: Jeff Masters , 11:58 AM GMT on May 17, 2015

Super Typhoon Dolphin intensified into a Category 5 storm with 160 mph winds and a central pressure of 925 mb at 2 pm EDT Saturday May 16, becoming Earth's fifth Category 5 storm of the year. Dolphin hung on to Category 5 strength for twelve hours before increasing wind shear helped knock the storm down to a Category 4 storm with 150 mph winds by Sunday morning. The eye of Dolphin passed through the channel between the islands of Guam and Rota Friday morning when the storm was at Category 2 strength with sustained winds of 110 mph. Guam experienced the weaker southern eyewall, and Rota saw the stronger northern eyewall. Andersen Air Force Base on Guam experienced sustained winds as high as 84 mph at 7:55 pm local time Friday, with a peak gust of 106 mph. Rainfall amounts tallied 9.30" in a 12-hour period. Dolphin knocked out power and damaged some homes, but the islands escaped serious destruction; Guam and Rota were very lucky that Dolphin waited a day to put on its intensification into a Category 5 storm. Satellite loops show a very impressive, well-organized system with a large ring of intense eyewall thunderstorms with very cold clouds tops, and prominent 28-mile diameter eye. A strong trough of low pressure has recurved the storm to the north, and Dolphin may pass close enough to Iwo Jima on Tuesday to bring that island typhoon conditions.


Figure 1. Still frame from 05 UTC May 16 2015, taken from a remarkable super high-resolution (every 2.5 minutes at 0.5 km) visible satellite loop of Super Typhoon Dolphin on Saturday morning, May 16, put together by The University of Wisconsin CIMSS group, using imagery from the new Japanese Himawari-8 satellite. At the time, Dolphin was an intensifying Category 4 super typhoon with 150 mph winds. More info and hi-res infrared loops can be found at the CIMSS satellite blog.

An unusual number of Category 5 storms so far in 2015
May 16 is exceptionally early to be getting our third Category 5 storm of the year in the Northwest Pacific. The global record for Category 5 storms is held by the El Niño year of 1997, which had twelve Category 5 storms--ten of them in the Northwest Pacific. The third Cat 5 of 1997 in the Northwest Pacific occurred on July 22, so we are more than two months ahead of that year's record pace. Dolphin is also the earliest-appearing 7th named storm of the Northwest Pacific's typhoon season; the previous record was on May 19, 1971. Super Typhoon Dolphin is already Earth's fifth Category Five storm this year, which is an unusually large number of these high-end tropical cyclones for so early in the year. Earth averaged just 4.6 Category 5 storms per year between 1990 - 2014, so we've already exceeded our average for an entire year; 2015 already has the 6th most Category 5 storms for any year in the past 26 years (reliable satellite records of Southern Hemisphere tropical cyclones extend back to 1990, so we only have about a 26-year period of decent records for global Category 5 tropical cyclones.) The majority of these storms occur during the July - November peak of the Northern Hemisphere's tropical cyclone season, with 59% of all Cat 5s occurring in the Northwest Pacific, so it is likely we will see several more Cat 5s this year. The early and violent start to 2015 typhoon season is due, in part, to exceptionally warm ocean temperatures in the typhoon breeding region between 5 - 10°N near the Date Line. These temperatures have been over 2°C (3.6°F) above average in recent months, due to a strengthening El Niño event.

The other four Category 5 storms of 2015: Note that Wikipedia lists the Southwest Indian Ocean's Tropical Cyclone Bansi as being a Category 5, but the Joint Typhoon Warning Center (JTWC) did not, giving it a top top wind speed of 150 mph, below the 156 mph threshold for Category 5. The JTWC is used here to rate Category 5 storms.


Cat Five #1, Southwest Indian. Tropical Cyclone Eunice as seen by the MODIS instrument at 05:30 UTC January 30, 2015, when the storm was at peak intensity (160 mph winds, 900 mb pressure.) Eunice was the 3rd strongest cyclone ever observed in the Southwest Indian Ocean by pressure, and the strongest by winds. Fortunately, Eunice affected only ocean areas in the South Indian Ocean.


Cat Five #2, Southeast Pacific. Tropical Cyclone Pam near peak intensity (165 mph winds, 896 mb pressure), as seen by the VIIRS instrument on the Suomi satellite at 10:42 am EDT March 13, 2015, just southeast of Efate Island, where the capital of Vanuatu, Port Vila, lies. Pam killed 16 people and did $250 million in damage to the island nation of Vanuatu, making it the 2nd most expensive disaster in their history (the most expensive, according to EM-DAT: Cyclone Eric of 1985, which did $173 million in damage in 1985 dollars, or $377 million 2015 dollars.) Pam was the third most intense storm in the entire Southern Hemisphere by central pressure, only after Cyclone Zoe of 2002 and Cyclone Gafilo of 2004. Pam is tied with Cyclone Orson and Cyclone Monica for having the strongest sustained 10-minute average winds of any cyclone on record in the Southern Hemisphere. Image credit: @NOAASatellites.


Cat Five #3, Northwest Pacific. Super Typhoon Maysak as seen from the International Space Station at approximately 6 pm EDT Tuesday March 31, 2015 (just after dawn local time.) At the time, Mayask was at peak intensity, with sustained winds of 160 mph (as estimated by the Joint Typhoon Warning Center) and a central pressure of 905 mb (as estimated by the Japan Meteorological Agency.) Maysak was the strongest typhoon ever observed in the Northwest Pacific prior to April, and one of only three Category 5 typhoons ever observed in the Northwest Pacific so early in the year. Maysak killed 9 and did $8.5 million in damage to several small islands in the Federated States of Micronesia, which it struck at Category 5 strength on March 31. The nine people killed by the storm made it Micronesia's second deadliest storm in recorded history, according to EM-DAT. Their deadliest disaster was Category 4 Typhoon Chataan, which dumped 19.90" (506 mm) of rain in 24 hours on Chuuk, causing landslides that killed 47 people. Maysak is the 2nd most expensive disaster in Micronesia's history; the most expensive was Category 1 Typhoon Nina, which did $6 million (1987 dollars) in damage on November 21, 1987.


Cat Five #4, Northwest Pacific. Super Typhoon Noul as seen by the new Japanese Himawari-8 satellite at 00:50 UTC May 10, 2015. At the time, Noul was a peak-strength Category 5 storm with 160 mph winds and a 915 mb central pressure. Noul hit northeast Luzon in the Philippines on May 10, killing two and causing less than $1 million in damage. Noul was the third strongest typhoon on record for so early in the year. The only stronger ones were Super Typhoon Amy of 1971, which deepened to 890 mb on May 2, and Super Typhoon Iris of 1951, which hit 909 mb on May 3. Image credit: The University of Wisconsin/CIMSS, which put together a remarkable hi-res satellite animation of the storm from the Himawari-8 satellite (which is still in check-out mode.)

Jeff Masters

Hurricane

Severe Weather Returns to the Plains This Weekend

By: Bob Henson , 5:45 PM GMT on May 15, 2015

Supercells may again roam the southern and central Great Plains on Friday and Saturday, as they did last weekend. The trigger comes from a strong, cold upper-level low that’s delivering an unseasonably late dose of rain to parts of drought-plagued southern California. Major League Baseball was halted on Thursday night by the first rain delays in four years at San Diego and the first in 11 years in Los Angeles. Following 0.55” of rain for the month up through Wednesday, San Diego’s Lindbergh Field scored 1.51” in less than two hours on Thursday evening. The calendar-day total of 1.63” was a record for any day in May, and as of 10:00 am PDT Friday, Lindbergh Field has received 2.25” for the month thus far, with rain continuing. In records going back to 1850, the wettest May in San Diego was 2.54” in 1921.

Much like last week’s upper low, the current one will turn sharply northward as it pushes onto the plains on Saturday. The slow eastward progress, coupled with ample low-level moisture from the Gulf of Mexico, will foster multiple rounds of showers and thunderstorms, with cells darting northward as the entire system moves slowly eastward. Flash flooding may again be a threat, especially in places like far north-central Texas and central Oklahoma that are already waterlogged from upwards of 10” of rain over the last two weeks.


Figure 1. A strong upper low at the 300-mb level (about 30,000 feet) high will be centered over Nevada at 0000 GMT Saturday, according to the 0000 GMT Friday run of the GFS model. The low will bring jet-stream-level winds approaching 100 knots (115 mph) across parts of Kansas, Oklahoma, and Texas on Saturday, boosting the odds of severe weather. Image credit: WunderMap.

Friday’s action should be focused near a developing surface low in far northeast Colorado, sweeping into eastern Wyoming and western Nebraska. These storms may quickly turn severe by afternoon, but should affect only sparsely populated areas. The late arrival of the upper low may also trigger an isolated storm or two further south along a dry line in far western Kansas. Tornadoes are possible, but the more serious tornado threat will likely be on Saturday, as stronger high-level winds associated with the upper low boost the vertical wind shear across Kansas and Oklahoma. The latest outlooks from NOAA’s Storm Prediction Center show large slight-risk areas for both Friday and Saturday, with enhanced-risk pockets centered in western Nebraska on Friday and western KS/OK on Saturday.

The ingredients are lining up for a potential major outbreak on Saturday evening, but it’s possible that morning showers and storms could keep the pot from boiling. During some years, spring thunderstorms are strongly “capped” across the Southern Plains by a recurrent layer of very warm air one to two miles above the surface that moves across the region from New Mexico and west Texas. This spring, a strong low-latitude jet stream (likely aided by El Niño) has led to cool and wet conditions across NM and west TX and cut back on the strength of the cap. This allows showers and storms to develop more readily over the Plains, but it also hinders the buildup of explosive instability that can enhance the severity of late-day storms. A large complex of morning thunderstorms cooled the airmass and largely quashed severe weather last Saturday across much of Oklahoma and southern Kansas. It’s possible a similar scenario will play out again this Saturday, as suggested by the 1200 GMT Friday runs of both the GFS and NAM models. If the air mass does manage to warm and destabilize, there could be a significant tornado threat on Saturday afternoon in western KS and OK, heading toward the Oklahoma City and Wichita areas by evening.




New this month: Community-inspired WunderPosters
Many thanks to those of you who submitted images and ideas for the community-inspired phase of our ongoing WunderPoster series. We sifted through hundreds of submissions to make our final selections, and the first two installments created by the WU design team are now online!

Roll clouds sometimes appear this time of year when the cool outflow around large thunderstorm complexes encounters warm, moist air. A sea breeze can serve the same function as the outflow, so roll clouds may also be spotted near coastal areas—but they’re not extremely common, so catching one is a rare treat. This poster was designed by Jerimiah Brown and inspired by a photo submitted by @The_Yarniverous.

Some of the most delicate beauty of wintertime comes to us in the form of hoarfrost. The name is derived from an old English word that alludes to the white hair earned by age. If a night is especially clear, still, and cold, then water vapor may be deposited in crystalline form directly onto twigs, wires, and other surfaces, sometimes branching into intricate structures. An image submitted by @bee.herder served as the inspiration for this poster, which was designed by Lauren Moyer.

Each of the 15 WunderPosters produced to date can be downloaded in formats suitable for posters or postcards.

Have a great weekend, everyone!

Bob Henson


Figure 2. A classic thunderstorm-generated roll cloud rolls across western Kansas near Gem on June 26, 2014. “Not even the panorama did it justice...it was a true 180!” Image credit: wunderphotographer clkngrny.


Figure 3. Pine needles are coated with hoarfrost on February 7, 2014, in Zaporizhia, Ukraine. Image credit: wunderphotographer TanyaMass.

Tornado

Category 2 Typhoon Dolphin Hits Guam

By: Jeff Masters , 11:54 AM GMT on May 15, 2015

Typhoon Dolphin blasted the islands of Guam and Rota in the U.S. Mariana Islands today as a Category 2 storm with sustained winds of 110 mph. The eye of Dolphin passed through the channel between the islands of Guam and Rota near 5 am EDT Friday, with Guam experiencing the weaker southern eyewall, and Rota seeing the stronger northern eyewall. Andersen Air Force Base on Guam experienced sustained winds as high as 84 mph at 7:55 pm local time; a peak gust of 106 mph occurred at 6:58 pm. Rainfall amounts tallied 9.30" in a 12-hour period. No damage reports have come out of Guam yet, but the damage is likely to be modest, given that Dolphin stayed below Category 3 strength, and the stronger northern eyewall missed the most heavily populated island in the chain--Guam. The 2 am EDT Friday advisory from the Joint Typhoon Warning Center (JTWC) put Dolphin's winds at 110 mph, just below the Category 3 threshold of 115 mph, and the Japan Meteorological Agency estimated that the central pressure had held steady at 955 mb. The lack of intensification was due to the fact that wind shear has been in the moderate range, 10 - 20 knots, driving dry air into the circulation. Satellite loops on Friday morning showed that Dolphin had changed little over the past 24 hours, and no eye was apparent. Sea surface temperatures (SSTs) were near 29°C (84°F), and warm waters extend to great depth along Dolphin's track, giving it plenty of heat energy to draw upon for intensification. Dolphin should be able to intensify to Category 4 typhoon status over the weekend as it takes advantage of lower shear and improved upper-level outflow. A strong trough of low pressure will recurve the storm to the north, and Dolphin may pass close enough to Iwo Jima on Tuesday to bring that island typhoon conditions.

The GFS model is no longer forecasting that a tropical disturbance near the Equator in the waters southeast of Guam (95W) will organize into a tropical depression.


Figure 1. Typhoon Dolphin as seen from the Guam radar just before it failed. The eye of Dophin was in the channel between the islands of Guam and Rota, with Guam experiencing the weaker southern eyewall, and Rota seeing the stronger northern eyewall.


Figure 2. Dolphin chews Guam: Typhoon Dolphin over Guam near 03 UTC May 15, 2015, as seen by the MODIS instrument on NASA's Aqua satellite. Image credit: NASA Worldview.

The last typhoon on Guam: thirteen years ago
As discussed in detail in Wednesday's post, the last typhoon to strike Guam was Typhoon Pongsona, which hit the island as a Category 4 super typhoon with 150 mph winds on December 8, 2002. The last tropical storm to affect Guam was Tropical Storm Saomai of August 2006, which had 50 mph winds when it moved over the island. May is exceptionally early for Guam to be worrying about a typhoon; according to NOAA's Historical Hurricane Tracks website, no typhoon has affected the island in the months of February through June since record keeping began in 1945.


Video 1. Storm chaser Jim Edds is on Guam, and has been documenting the impact of Dolphin on the island via his Twitter feed with videos like this one.

The NWS in Guam is putting out special advisories and local statements on Dolphin.

A remarkable super high-resolution (every 2.5 minutes at 0.5 km) visible satellite loop of Typhoon Dolphin at sunrise on May 15 has been put together by The University of Wisconsin CIMSS group, using imagery from the new Japanese Himawari-8 satellite. A larger-scale view is available from NOAA, resembling a witch's cauldron full of boiling, poisonous brew! More info and hi-res infrared loops can be found at the CIMSS satellite blog.

Later today, Bob Henson will have an update on the severe weather potential for the U.S. on Friday and Saturday.

Jeff Masters

Hurricane

A Rare Mid-Year El Niño Event Is Strengthening

By: Bob Henson , 6:12 PM GMT on May 14, 2015

The robust El Niño event anticipated for more than a year is finally coming to fruition, according to the latest observations and forecasts. NOAA's latest monthly analysis, issued on Thursday morning, continues the El Niño Advisory already in effect and calls for a 90% chance of El Niño conditions persisting through the summer, with a greater-than-80% chance they will continue through the end of 2015. These are the highest probabilities yet for the current event, and a sign of increased forecaster confidence--despite the fact that we're in northern spring, the very time when El Niño outlooks are most uncertain.


Figure 1. A schematic showing the processes involved in El Niño. The trade winds shown by the arrow are weaker at this point than during La Niña or neutral conditions; at certain times and at some locations, they may even reverse, blowing from west to east. For a full explanation of the El Niño-Southern Oscillation, including additional graphics, see the website published by the Australian Bureau of Meteorology. Image credit: BOM.


Forecasters and computer models alike have been confounded by this event. In a classic El Niño, the ocean and atmosphere are synchronized in a mutually reinforcing pattern that pushes warm sea-surface temperatures (SSTs) and thunderstorm activity along the equator eastward for thousands of miles, from Indonesia toward South America (see Figure 1). Sometimes the atmosphere doesn't respond to a "kick" from the ocean, and an embryonic El Niño fails to develop. This was the case last spring, when a powerful oceanic Kelvin wave (a broad, shallow, slow-moving impulse) pushed warm water east across the Pacific tropics. Keying off this wave, many of the global models used in El Niño prediction called for a moderate or even strong El Niño by the fall of 2014. However, the normal east-to-west trade winds never reversed, which helped torpedo the needed ocean-atmosphere synchrony. The ocean tried again last fall with another Kelvin wave, but again the atmosphere failed to respond, and the SST warming disappeared after a few weeks.


Figure 2. Recent weekly departures from normal across the four tropical Pacific regions (top map) that are regularly monitored for signs of El Niño and La Niña. Image credit: NOAA Climate Prediction Center.


This time, things appear to be different: SSTs have warmed for the last several months, and more recently, trade winds have weakened. As of Monday, the weekly-averaged SSTs over the four regions monitored for El Niño were all at least 1.0°C above average (see Figure 2, right). If the values for all four regions can sustain this feat throughout the next month, it'll be the first time this has happened since November 1997, during the strongest El Niño event of the 20th century. Just as significant, the persistently warmer-than-normal SSTs over the western tropical Pacific have now cooled, which helps support the reversal of trade winds so critical to El Niño. The current SST map now resembles a more textbook-like El Niño signature (see Figure 3, below), and there is every indication that the ocean-atmosphere coupling will now continue to grow.

An event out of season
As far as the eastern tropical Pacific goes, it's beginning to look a lot like Christmas. El Niño--"the Christ child" in Spanish--gets its name from its tendency to bring above-average SSTs to the coasts of Peru and Ecuador around Christmastime. The climatology of the eastern Pacific tends to support El Niño and La Niña development during the northern autumn, maximum strength in the winter, and decay in the spring. The current event is thus bucking climatology as it continues into northern spring. The three-month departure from average in the Niño3.4 region reached the El Niño threshold of +0.5°C in Oct-Nov-Jan 2014-15, and it's hovered in the weak range (+0.5 to +1.0°C] ever since, with a value of +0.6°C for Feb-Mar-Apr 2015. Only 12 of the 65 prior years in the historical database of El Niño/Southern Oscillation (ENSO) events have seen a value of at least +0.5°C during the Feb-Mar-Apr period. Water temperatures in the Niño3.4 region are normally at their warmest in May, so the current warm anomaly is leading to especially toasty SSTs of around 29°C (84°F). If this El Niño event does intensify, as models strongly suggest it will (see below), it'll be one for the record books. There are no analogs in the database for a weak event in northern winter that becomes a stronger event by summer. Persisting into northern fall will also greatly raise the odds of this becoming a rare two-year event. In every case since at least 1950 when El Niño conditions were present by Jul-Aug-Sep, the event continued into the start of the next calendar year. Two-year El Niños are more unusual than two-year La Niñas, but they do happen, as in 1968-1970 and 1986-1988. See the new climate.gov blog by Emily Becker for more on the unusual timing of this event.


Figure 3. Departures from normal for sea-surface temperatures as measured on May 13, 2015. The warmer-than-average belt across the central and eastern tropical Pacific is characteristic of El Niño. Image credit: tropicaltidbits.com.


Northern spring is an especially difficult time to predict El Niño evolution. Computer-model skill at predicting ENSO is at its lowest then, in part because of reduced east-west gradients in SSTs across the tropical Pacific, but also due to factors that have yet to be fully understood. "The Spring Barrier is the climate forecaster’s equivalent of mayhem," says Michelle L'Heureux in an excellent climate.gov discussion of what forecasters often call the "spring predictability barrier." Skill does begin to improve for forecasts produced in May, according to L'Heureux, so we can begin placing more trust in the 2015-16 El Niño predictions from this point onward--although L'Heureux notes that even model runs produced in August still miss about a quarter of the winter variability in ENSO.

How strong will it get?
This week's Niño3.4 SST anomaly of +1.0°C is at the threshold of a moderate-strength event. Another 0.5°C would push the event into the strong range, which was last observed in late 2009 and early 2010. Klaus Wolter (NOAA Earth Systems Laboratory) has devised a Multivariate ENSO Index (MEI) that uses multiple indicators to diagnose El Niño and La Niña. Last year's event briefly nudged into Wolter's "strong" category (defined as the top 10% of events) before subsiding. The MEI is now again at the threshold of "strong," and a statistical model recently run by Wolter finds a 44% chance that strong conditions will be in place during Aug-Sep 2015, the time of year when this statistical model is most accurate. "We have had some pretty unusual (non-persistent) behavior of the ENSO-system in the last four years that was anticipated better than by flipping a coin, especially last year, but certainly not perfectly," said Wolter in an email update.


Figure 4. Projected values of the Niño3.4 departure from average by September, based on ensemble averages from a variety of global ocean-atmosphere models (listed along left-hand side). The La Niña and El Niño thresholds indicated by the shading on this graphic are 0.8°C, the values used in Australia. The comparable threshold used by NOAA is 0.5°C, because U.S. impacts can occur with smaller departures from average. Image credit: Australia Bureau of Meteorology.


The dynamical models run at various centers around the world to predict ENSO are now unanimous in keeping El Niño going into northern autumn. The values shown in Figure 4 (above) are ensemble averages for each model, which means they smooth out the range of outcomes depicted by multiple runs of the same model. (Each run has slight differences in its starting point, to account for features too small to be observed and the natural variations that result). Within each ensemble, there's a wide range of outcomes projected by autumn 2015, from a borderline El Niño to much more extreme values. Figure 5 (below) includes both the ensemble average and the individual members for the seven models in the North American Multi-Model Ensemble. A number of individual model runs push the Niño3.4 index well above +2.5°C over the next few months, and the entire NMME average is around +2.2°C for November and December. By comparison, the highest three-month departure observed in the entire 65-year NOAA database is +2.4°C (Nov-Dec-Jan 1997-98).

Now is a very good time to keep in mind that global models tend to hyperventilate a bit when it comes to strong ENSO events. "This is because the El Niño events are too shallow in the models," says Kevin Trenberth (National Center for Atmospheric Research). "They don't have as much ocean heat content engaged, so there is more of a surface signal." As for the stark variation among individual model runs, it may be due to the spring predictability barrier, as well as the result of another very powerful Kelvin wave and a strong westerly wind burst now traversing the Pacific. Models can easily predict a strengthening of El Niño conditions over the next several months as these features continue eastward, but it's tougher for the models to discern exactly what will happen after the Kelvin wave reaches South America. Trenberth points to the ocean-atmosphere coupling known as the Bjerknes feedback mechanism: "What happens after this Kelvin wave response is all over the place. This El Niño is being fought by the annual cycle, which tries to make SSTs cold by Sept-Oct.  That tendency keeps the warmest waters back near the International Date Line and cuts off the Bjerknes feedback.  If the SSTs develop to be big enough to overcome the annual cycle tendencies, then the Bjerknes feedback can kick in."


Figure 5. Projected evolution of Niño3.4 temperatures from members of the North American Mutli-Model Ensemble (members listed at top left). Dashed lines denote individual model runs; solid lines denote ensemble averages for each model. Image credit: NOAA Climate Prediction Center.


What can we expect this summer?
Because it's quite rare to have intensifying El Niño conditions at this time of year, it's difficult to glean a confident signal from past events on how El Niño might affect U.S. summer weather. The global effects of El Niño arise from a shifting of showers and thunderstorms into the central and eastern tropical Pacific, which causes a reverberating sequence of atmospheric waves that tend to enhance precipitation in some areas and reduce it in others. In midlatitudes, these relationships, called teleconnections, are usually strongest in the winter hemisphere; for example, Australia often falls into drought when El Niño is developing in Jun-Jul-Aug (see Figure 6). If a strong El Niño does develop and persists into northern winter, the likely U.S. impacts would be more clear-cut, including wetter-than-average conditions across the southern half of the country, from California through Texas to Florida. This month could be seen as a sneak preview of sorts, with soggy conditions prevalent across the central and southern Plains and two unusually-wet-for-May systems reaching southern California, one last weekend and another now on its way. There is some hope for drought relief in the Golden State, given that the odds of an wetter-than-normal California rise sharply for the strongest El Niño events, but by no means would a wet winter be guaranteed. The strong El Niño of 1987-88 (which happened to be the second year of a two-year event) produced a drier-than-average winter from California to Washington.

Given that El Niño tends to suppress hurricane formation in the North Atlantic, the odds of a quiet season in that basin are growing by the month. However, a season with few storms doesn't necessarily translate into a low-impact year: the anemic 1992 season included the catastrophic Hurricane Andrew. And it's possible (though unlikely) to have a busy Atlantic hurricane season even during El Niño. Right in the middle of the weak-to-moderate two-year El Niño event of 1968-70, the Atlantic produced its most active season in 36 years, with a total of 18 named storms, 12 hurricanes, and 5 major hurricanes--including the horrific Hurricane Camille.

Busy day on the blog!
For more on today's happenings, see the earlier posts by Jeff Masters on Typhoon Dolphin, which is threatening Guam, and on the incredibly hot temperatures in Spain, which have smashed European records for the month of May.

Bob Henson


Figure 6. Global ENSO teleconnections (seasonal tendencies linked to El Niño and La Niña) for northern summer (June - August). Image credit: NOAA.

El Niño

Typhoon Warnings on Guam for Category 2 Typhoon Dolphin

By: JeffMasters, 3:23 PM GMT on May 14, 2015

Typhoon warnings are flying for the U.S. Mariana Islands, including Guam, Rota, Tinian, and Saipan, as Typhoon Dolphin races west-northwest towards the islands at 18 mph. Guam radar shows the outer bands of Dolphin are already affecting the islands, and Guam is under a Flash Flood Watch for rainfall amounts of 8 - 12". Dolphin has intensified only slightly in the past 24 hours; the 8 am EDT Thursday advisory from the Joint Typhoon Warning Center (JTWC) put Dolphin's winds at 110 mph, up from 105 mph 24 hours previous to that, and the Japan Meteorological Agency estimated that the central pressure had held steady at 965 mb. The lack of intensification is due to the fact that wind shear has been in the moderate range, 10 - 20 knots, driving dry air to the west of Dolphin into the circulation. Satellite loops show that Dolphin has shown a slight increase in the intensity and areal coverage of its heavy thunderstorms on Thursday morning, but no eye was apparent. Sea surface temperatures (SSTs) are near 29°C (84°F), and warm waters extend to great depth along Dolphin's track, giving it plenty of heat energy to draw upon for intensification. Dolphin should be able to intensify to Category 3 typhoon status by the time it reaches Guam, but the continued presence of moderate wind shear and dry air should keep the storm below Category 4 strength until after it passes Guam. The latest 00Z Thursday run of the European model and 06Z run of the GFS model show Dolphin passing about 30 - 110 miles to the north of Guam between 06 - 09 UTC Friday (2 am - 5 am U.S. Eastern Daylight Time.) On this track, the island of Rota would receive the worst of Dolphin, and Guam would experience the weaker (left front) side of the typhoon. A strong trough of low pressure should recurve the storm to the north out to sea late this weekend, and Dolphin may pass close enough to Iwo Jima on Tuesday to bring that island typhoon conditions. The GFS model is also advertising that a tropical disturbance near the Equator in the waters southeast of Guam (95W) will organize early next week into a tropical depression, but it is too early to be confident of this prediction.


Figure 1. Latest image from the Guam radar.

The last typhoon on Guam: thirteen years ago
As discussed in detail in yesterday's post, the last typhoon to strike Guam was Typhoon Pongsona, which hit the island as a Category 4 super typhoon with 150 mph winds on December 8, 2002. The last tropical storm to affect Guam was Tropical Storm Saomai of August 2006, which had 50 mph winds when it moved over the island. May is exceptionally early for Guam to be worrying about a typhoon; according to NOAA's Historical Hurricane Tracks website, no typhoon has affected the island in the months of February through June since record keeping began in 1945.

Storm chaser Jim Edds is on Guam, and will be documenting the impact of Dolphin on the island via his Twitter feed.

The NWS in Guam is putting out special advisories and local statements on Dolphin. Wave heights are expected to peak at 22 - 30 feet as the typhoon passes the island.

Bob Henson will have an El Niño post by 2 pm EDT.

Jeff Masters

Hurricane

All-Time May Heat Record for Europe Falls For the 2nd Time This Month

By: Jeff Masters , 1:01 PM GMT on May 14, 2015

An extreme May heat wave unprecedented in European recorded history has invaded Spain and Portugal, bringing the hottest May temperatures ever recorded on the continent. According to the Spanish meteorological agency, AEMET, at least four stations in the Valencian Community of eastern Spain hit temperatures today in excess of the previous European May heat record set just eight days ago--a 41.9°C (107.4°F) reading at Catenanuova, Sicily (Italy) on May 6. Today's European record-breaking May temperatures in Spain included:

Carcaixent AEMET: 44.4°C (111.9°F)
Xativa: 42.7°C (108.9°F)
Algemesi: 42.6°C (108.7°F)
Valencia: 42.6°C (108.7°F)

Many stations in Spain's Valencian community went above their June records, and were near their all-time records for any month. The record set at Valencia Airport today was 6.6°C (11.9°F) above the previous highest May temperature, was 4.4°C (7.9°F) higher than the record for June, and was the 3rd hottest temperature since records began in 1869 for any month! This week's heat wave began yesterday, when hot air from North Africa flowed northwards over Spain and Portugal, setting all-time May heat records at Madrid, Sevilla, Cordoba, Ciudad Real, Granada, and many other cities. Portugal beat its all-time May heat record with a 40.0°C (104.0°F) reading at Beja EMA (old record: 39.5°C, 103.1°F, at Regua on May 28, 2001.) The most remarkable record yesterday, however, was from the Canary Islands to the southwest of Spain, where Lanzarote Airport hit 42.6°C (108.7°F), breaking its old record for the entire month of May by 6°C (10.8°F)! The old record was 36.6 °C (97.9°F) on May 24, 1986.


Figure 1. Predicted high temperatures for Thursday, May 14, 2015 for Spain from AEMET. Temperatures of 42.0°C are hotter than any May readings on record for Spain.

According to weather records researcher Maximiliano Herrera, Europe also had its earliest 40°C temperature ever recorded this month--a 40.4°C (104.7°F) reading at San Priamo in southeastern Sardinia on May 5. Prior to 2015, the previous all-time hottest May temperature recorded in Europe was 41.7°C (107.1°F), on May 17, 2006 in Andujar, Spain. What's really remarkable about the two European all-time May heat records set this month is that they have occurred in the first half of the month. We would expect all-time monthly heat records to fall near the end of the month, as summer's heat begins to really ramp up. Note that the World Meteorological Organization (WMO) does not maintain official monthly all-time extreme temperature records for the continents, so the records quoted here are from Maximiliano Herrera.

Thanks go to Maximiliano Herrera and Michael Theusner for the temperature records provided in this post.

Jeff Masters

Heat

Typhoon Dolphin Intensifying as it Heads Towards Guam

By: Jeff Masters , 3:37 PM GMT on May 13, 2015

Typhoon Dolphin is gathering strength in the waters to the east of the U.S. Mariana Islands, and is a serious threat to strike Guam and the nearby islands as a major typhoon on Friday morning (U.S. EDT.) The 8 am EDT Wednesday advisory from the Joint Typhoon Warning Center (JTWC) put Dolphin's winds at 105 mph, and the Japan Meteorological Agency estimated a central pressure of 965 mb. The latest 00Z Wednesday run of the European model and 06Z run of the GFS model show Dolphin passing within 50 miles of Guam between 06 - 09 UTC Friday (2 am - 5 am U.S. Eastern Daylight Time.) Satellite loops show that Dolphin significantly increased in organization on Wednesday morning, with an increase in the intensity and areal coverage of its heavy thunderstorms and formation of an eye. Sea surface temperatures (SSTs) are near 29°C (84°F), and warm waters extend to great depth along Dolphin's track, giving it plenty of heat energy to draw upon for intensification. Wind shear has risen over the past day to the moderate range, 10 - 20 knots, and is expected to remain low to moderate through Friday. Dolphin should be able to intensify to Category 3 typhoon status by the time it reaches Guam, and may rapidly intensify, potentially affecting Guam as a Category 4 super typhoon. Guam will likely be the last land area Dolphin will affect, as a strong trough of low pressure should recurve the storm to the north out to sea late this weekend. The GFS model is also advertising that a tropical disturbance near the Equator in the waters southeast of Guam (95W) will organize early next week into a tropical depression, but it is too early to be confident of this prediction.


Figure 1. Typhoon Dolphin as seen by the MODIS instrument on NASA's Aqua satellite at approximately 03 UTC May 13, 2015. At the time, Dolphin had sustained winds of 85 mph, and a "pinhole" eye with a diameter of only six miles. Image credit: NASA Worldview.

The last typhoon on Guam: thirteen years ago
The last typhoon to strike Guam was Typhoon Pongsona, which hit the island as a Category 4 super typhoon with 150 mph winds on December 8, 2002. Sustained winds of 144 mph with gusts to 173 mph were recorded on the island, and Andersen Air Force Base was in the eye for two hours. The lowest pressure on Guam was 935 millibars, making Pongsona the third most intense typhoon to strike the island (the others: a 1900 typhoon with a 926 mb pressure, and Typhoon Karen of 1962, at 932 mb). With strong building standards and experience from repeated typhoon strikes (six typhoons in the previous ten years), there was one death from flying glass, and 193 injuries. Damage was over $700 million (2002 US$, $918 million 2015 USD), making Pongsona among the five costliest typhoons on the island. The typhoon also caused heavy damage on Rota and elsewhere in the Northern Mariana Islands, and as a result of its impact, the name Pongsona was retired. The last tropical storm to affect Guam was Tropical Storm Saomai of August 2006, which had 50 mph winds when it moved over the island.


Figure 2. Super Typhoon Pongsona as seen on December 8, 2002. At the time, Pongsona was at its peak strength--Category 4 with 150 mph winds. The image shows Pongsona over the Northern Mariana Islands in the western Pacific Ocean approaching Guam. Image credit: Jacques Descloitres, MODIS Rapid Response Team, NASA/GSFC - http://visibleearth.nasa.gov/view_rec.php?id=4761

A record early start to typhoon season
May is exceptionally early for Guam to be worrying about a typhoon; according to NOAA's Historical Hurricane Tracks website, no typhoon has affected the island in the months of February through June since record keeping began in 1945 (one January storm, Typhoon Roy of 1988, did pass near Guam in January, though.) Guam's early typhoon worries this year reflect how crazy-busy the early part of the 2015 typhoon season has been. The Japan Meteorological Agency (JMA) gave Tropical Storm Dolphin its name on May 9. According to statistics from the Japan Meteorological Agency's database from 1951 - 2015 maintained by Digital Typhoon, this was the earliest appearance on record of the Northwest Pacific's seventh named storm of the year. The previous record was May 19, 1971 (Carla.) Usually by this time of year, just two named storms have appeared; the seventh storm of the year typically doesn't form until the third week of July. According to Colorado State University hurricane scientist Phil Klotzbach, the seven storms so far in 2015 have been unusually strong: Northwest Pacific Accumulated Cyclone Energy (ACE) was at record high through May 11 (78.1), just ahead of the old record of 77.7 in 1957. The early and violent start to 2015 typhoon season is due, in part, to exceptionally warm ocean temperatures in the typhoon breeding region between 5 - 10°N near the Date Line. These temperatures have been over 2°C (3.6°F) above average in recent months, due to a strengthening El Niño event.

Storm chaser Jim Edds is on Guam, and will be documenting the impact of Dolphin on the island via his Twitter feed.

Jeff Masters

Hurricane

Florida Has its Warmest April on Record

By: Jeff Masters , 11:23 PM GMT on May 12, 2015

Florida residents, if you thought April 2015 seemed ridiculously hot, you were right--April 2015 was Florida's hottest April on record, said NOAA's National Climatic Data Center (NCDC) on Tuesday. The contiguous U.S. experienced its 17th warmest April since 1895, with the most notable heat in the Southeast U.S., where Georgia and South Carolina also experienced top-ten warmest Aprils on record. No portions of the country were notably cooler than average. The year-to-date period January - April ranked as the 20th warmest year-to-date on record. It was a relatively wet month, ranking in the wettest one-third of the historical record, and portions of the Central U.S. experienced notable relief from drought conditions. The U.S. Climate Extremes Index (USCEI) for the year-to-date was 40 percent above average and the 15th highest value since 1900. The warm West and cold Northeast temperature pattern during January-April contributed to the much above average USCEI, with the components that measure both warm and cold daytime and nighttime temperatures being much above average. The USCEI is an index that tracks extremes (falling in the upper or lower 10 percent of the record) in temperature, precipitation and drought across the contiguous United States


Figure 1. Statewide temperature rankings for April 2015. Florida had its hottest April on record, and Georgia and South Carolina also had top-ten warmest Aprils. Cool weather was noticeably absent. Image credit: NOAA/NCDC.

Record heat in Florida, Cuba, and Puerto Rico
According to NOAA, April 2015 was the hottest April on record at both Miami and Fort Lauderdale, and the second hottest at West Palm Beach and Naples. Almost every day in April had warmer-than-normal temperatures, with no more than three days of cooler-than-normal temperatures at any of the main climate sites. The heat peaked on April 26 with the hottest day in southeast Florida in almost six years, when high temperatures reached the mid to upper 90s at official recording stations all across the area. One official station at Royal Palm Ranger Station in the Everglades hit 100 degrees, the first time this has happened in South Florida in the month of April. The heat was even more impressive in nearby Cuba, where Havana set its all-time temperature record that same day with a sizzling 37.0°C (98.6°F); Holguin had the second-highest temperature ever recorded anywhere in Cuba: 38.7°C (101.6°F). The main cause of the April heat was a persistent and strong high pressure area which not only brought warm, subtropical air into the region but kept cold fronts from making southward progress down the state. April's weather was more like what normally occurs in June, and the hot and windless conditions helped warm up the waters where Tropical Storm Ana formed, leading to the 2nd earliest landfalling tropical storm in recorded U.S. history.

Remarkable heat also affected the Caribbean during the last week of April, with San Juan, Puerto Rico setting daily record highs on six consecutive days, April 25 - 30. This ties their record for most consecutive days with a record high, set June 21 - 26, 1983. San Juan hit 94°F four times during the last week of April; prior to this April there were only twelve times in recorded history that the temperature hit 94 or above in April (including two readings of 95 and one of 97.)

Wunderblogger Steve Gregory discusses how a typical late spring weather pattern has emerged over the U.S. in his Monday afternoon post.

Jeff Masters

Climate Summaries

Intensifying Tropical Storm Dolphin a Threat to Guam

By: Jeff Masters , 2:12 PM GMT on May 12, 2015

In the Northwest Pacific, Tropical Storm Dolphin is gathering strength in the waters to the east of the U.S. Mariana Islands, and is a threat to bring typhoon conditions to those islands--including Guam--on Thursday. The 8 am EDT Tuesday advisory from the Joint Typhoon Warning Center (JTWC) put Dolphin's winds at 60 mph, and JTWC predicted the storm would reach Category 1 typhoon strength by Wednesday morning (EDT.) The latest 00Z Tuesday run of the European model and 06Z run of the GFS model show Dolphin passing within 50 miles of Guam as an intensifying typhoon sometime between 21 UTC Thursday and 06 UTC Friday (Thursday night U.S. Eastern Daylight Time.) Satellite loops show that Dolphin significantly increased in organization on Tuesday morning, with more low-level spiral banding and an increase in the intensity and areal coverage of its heavy thunderstorms. Sea surface temperatures (SSTs) are near 29°C (84°F), and warm waters extend to great depth along Dolphin's track, giving it plenty of heat energy to draw upon for intensification. Wind shear has fallen to the low range, 5 - 10 knots, and is expected to remain low through Friday. Dolphin should be able to intensify to Category 2 typhoon status by the time it reaches Guam, and may rapidly intensify, potentially affecting Guam as a major Category 3 or stronger typhoon. Guam will likely be the last land area Dolphin will affect, as a strong trough of low pressure should recurve the storm to the north out to sea late this weekend. The GFS model is also advertising that a new tropical storm will form this weekend in the waters east of Guam, but it is too early to be confident of this prediction.


Figure 1. Tropical Storm Dolphin as seen by the MODIS instrument on NASA's Aqua satellite at 02:50 UTC May 12, 2015. At the time, Dolphin had sustained winds of 50 mph. Image credit: NASA.

The last typhoon on Guam: thirteen years ago
The last typhoon to strike Guam was Typhoon Pongsona, which hit the island as a Category 4 super typhoon with 150 mph winds on December 8, 2002. Sustained winds of 144 mph with gusts to 173 mph were recorded on the island, and Andersen Air Force Base was in the eye for two hours. The lowest pressure on Guam was 935 millibars, making Pongsona the third most intense typhoon to strike the island (the others: a 1900 typhoon with a 926 mb pressure, and Typhoon Karen of 1962, at 932 mb). With strong building standards and experience from repeated typhoon strikes (six typhoons in the previous ten years), there were no direct deaths from Pongsona's strike--but one indirect death from flying glass, and 193 injuries. Damage was over $700 million (2002 US$, $918 million 2015 USD), making Pongsona among the five costliest typhoons on the island. The typhoon also caused heavy damage on Rota and elsewhere in the Northern Mariana Islands, and as a result of its impact, the name Pongsona was retired. The last tropical storm to affect Guam was Tropical Storm Saomai of August 2006, which had 50 mph winds when it moved over the island.


Figure 2. Extensive damage on Guam caused by Typhoon Pongsona December 8, 2002. Image credit: Andrea Booher, U.S. Army Corps of Engineers - U.S. Army Corps of Engineers Digital Visual Library.

A record early start to typhoon season
May is exceptionally early for Guam to be worrying about a typhoon; according to NOAA's Historical Hurricane Tracks website, no typhoon has affected the island in the months of February through June since record keeping began in 1945 (one January storm, Typhoon Roy of 1988, did pass near Guam in January, though.) Guam's early typhoon worries this year reflect how crazy-busy the early part of the 2015 typhoon season has been. The Japan Meteorological Agency (JMA) gave Tropical Storm Dolphin its name on May 9. According to statistics from the Japan Meteorological Agency's database from 1951 - 2015 maintained by Digital Typhoon, this was the earliest appearance on record of the Northwest Pacific's seventh named storm of the year. The previous record was May 19, 1971 (Carla.) Usually by this time of year, just two named storms have appeared; the seventh storm of the year typically doesn't form until the third week of July. According to Colorado State University hurricane scientist Phil Klotzbach, the seven storms so far in 2015 have been unusually strong: Northwest Pacific Accumulated Cyclone Energy (ACE) was at record high through May 11 (78.1), just ahead of the old record of 77.7 in 1957. The early and violent start to 2015 typhoon season is due, in part, to exceptionally warm ocean temperatures in the typhoon breeding region between 5 - 10°N near the Date Line. These temperatures have been over 2°C (3.6°F) above average in recent months, due to a strengthening El Niño event. 

Storm chaser Jim Edds is on Guam, and will be documenting the impact of Dolphin on the island via his Twitter feed.

Jeff Masters

Hurricane

A Quiet Tornado Season Revs Up; Tropics Calm Down

By: Bob Henson and Jeff Masters , 4:29 PM GMT on May 11, 2015

After making landfall at 6:15 am EDT Sunday on the coast of South Carolina just south of the North Carolina border as the strongest tropical storm ever recorded to hit the U.S. so early in the year, Tropical Storm Ana dissipated on Sunday evening over North Carolina. The storm brought a number of wind gusts of 50 - 60 mph to the coast, but no significant flooding or damage was reported. Before Ana, only six tropical cyclones tracked over the Gulf Coast or East Coast before June 1. Ana was the earliest. (Tropical cyclones include tropical depressions, tropical storms and hurricanes.)


Figure 1. Infrared VIIRS image of Super Typhoon Noul as seen by Suomi satellite at 05:25 UTC May 10, 2015. At the time, Noul was nearing landfall on the northeastern tip of the Philippines' Luzon Island a Category 5 storm with 160 mph winds. Image credit: Lan Lindsey, NOAA/NASA RAMMB/CIRA.

Typhoon Noul (called Dodong in the Philippines) weakened to a Category 1 storm with top winds of 90 mph Monday morning after belting the extreme northeastern tip of the Philippines' Luzon Island as a Category 5 storm with 160 mph winds on Sunday morning. Two deaths are being blamed on the storm, but damage was limited by the fact that the northeastern Luzon is relatively sparsely populated. Satellite loops on Monday morning showed that Noul was barely recognizable as a typhoon, looking like a large misshapen blob of heavy thunderstorms. Noul is caught in a trough of low pressure that will swing the storm to the north and northeast, out to sea, skirting the Ryukyu Islands and the coast of Japan. The Yaeyama island chain of far southern Japan received some powerful winds from Noul on Monday morning; the Ishigaki-jima observation site--with records since 1897--set an all-time sustained wind record for the month of May, with 28.8 m/s (64.4 mph). Gusts up to 44.1 m/s (98.6 mph) also occurred. Farther east, Shimoji clocked a sustained wind of 36.4 m/s (81.4 mph) and a gust to 47.8 m/s (106.9 mph) around 1 a.m. JST (noon U.S. EDT). Thanks go to TWC's Nick Wiltgen for these wind stats.

Next up is Tropical Storm Dolphin, which formed over Micronesia on May 9, breaking the record set on May 19, 1971 for the earliest formation of the Northwest Pacific's seventh named storm of the year, according to statistics from the Japan Meteorological Agency's database from 1951 - 2015 maintained by Digital Typhoon. Dolphin is not yet well-organized, as seen on satellite loops, due to high wind shear. However, wind shear will relax to the low range by Wednesday, which should allow the storm to intensify into a typhoon when it makes its closest approach to Guam on Thursday and Friday. Guam will likely be the last land area Dolphin will affect, as a strong trough of low pressure should recurve the storm to the north out to sea late this week.

Tornadoes strike eastern South Dakota, east Texas
What began as one of the least active tornado seasons on record kicked into high gear over the weekend. Sunday produced a total of 26 tornado reports, following 53 reports on Saturday, according to preliminary data from NOAA's Storm Prediction Center. The worst strikes were early and late on Sunday. From around 10:30 to 11:00 am CDT, a tornado with a preliminary EF2 rating carved a quarter-mile-wide, 17-mile-long path across eastern South Dakota, causing 9 injuries and damaging more than 20 buildings in the town of Delmont. Another tornado struck Van, TX (about 100 miles east of Dallas) at around 8:45 pm CDT. The town's emergency management coordinator, Chuck Allen, estimated early on Monday morning that about 30 percent of the town of Van had been damaged, with some homes destroyed. At least 26 people were injured, with search and recovery operations continuing today.


Figure 2. Rev. David Otten, from Immanuel Lutheran Church in Dimock, SD, walks down Main Street in Delmont, SD, in front of Zion Lutheran Church after a tornado heavily damaged the structure on Sunday, May 10. Otten was filling in for Rev. Brian Bucklew but wasn't in the church when the tornado hit. Image credit: Joe Ahlquist/The Argus Leader, via AP.

The severe weather outbreak that began on Wednesday in Texas and Oklahoma will be winding down on Monday, as the upper-level low and associated cold front move into the Great Lakes and lower Mississippi Valley. Instability and wind shear have both decreased, but tornadic supercells are still possible, particularly in eastern Indiana, Ohio, and southeast Michigan. Heavy rain, wind, and hail can be expected along the length of the front, especially in Mississippi, Louisiana, and south Texas.


Figure 3. Preliminary severe weather reports for Saturday, May 9 (top) and Sunday, May 10 (bottom) show the bifurcated nature of the weekend outbreak, with one focal point near the surface low, which shifted from Colorado and Kansas to South Dakota and Iowa. Another focal point, along the trailing cold front, remained in Texas and Oklahoma. Image credit: NOAA/SPC.


After today's storms, the nation will see a brief respite, but forecast models strongly suggest that another multiday outbreak will unfold over the coming weekend, likely starting in Colorado and Kansas on Friday and expanding further into the Great Plains on Saturday. Thus far we've seen about 150 preliminary tornado reports for May, compared to an average for the entire month of 173 during the last three years (2012 - 2014). After Sunday's tornadoes, our preliminary total for the year is now around 400, just below the "inflation-adjusted" average for this point in the year. The average year-to-date total in that database climbs to around 600 by May 31. It would take a very busy last half of May to reach that point, but given the model indications for this coming weekend, and the longer-range suggestions of continued upper-level troughiness in the West, I wouldn't rule it out.

The good news is that tornado deaths and injuries have been remarkably low to date. Only three tornado-related fatalities appear in the SPC database through May 7. The years 2012 through 2014 averaged 34 deaths by the end of May, and the year before that--2011--saw more than 500 deaths by that point. The relatively light human toll is partially due to luck (this year's tornadoes have been missing large population centers) and partially due to the very small number of stronger tornadoes. Only one EF4 tornado and three EF3 tornadoes had been confirmed in unofficial totals through May 6. Over the last several decades, the total number of EF3/EF4/EF5 tornadoes has ranged between about 20 and 60, with 2011 a notable exception (more than 80 were observed that year). Most of these are rated EF3; only a handful of EF4 tornadoes occur in a typical year, and only about a dozen EF5s can be expected per decade.


Figure 4. This tornado northeast of Eads, CO, on Saturday, May 9 is approaching the end of its life cycle. Image credit: Bob Henson.

Snow and tornadoes in the same state--on the same day?
Old-timers from Colorado to South Dakota were impressed but not shocked by the wind-driven, late-season snowfall that wrapped around the back side of this weekend's strong low-pressure system over the Plains. Rapid City has recorded measurable snow as late as June 13, 1969, and Denver averages 1.7" in May (albeit with large year-to-year variability). Ground temperatures are warming up quickly, so to get much accumulation, the snow has to be falling heavily. Likewise, in order to cool down an already-chilly airmass enough to make it snow this late in the spring, heavy precipitation is usually needed. The Rapid City NWS office reported 13.6" on Saturday and Sunday, its second heaviest May snowstorm and the heaviest ever recorded so late in the season. Denver's totals of 3.5" on Saturday and 0.5" on Sunday were both enough to set daily records, but the month's all-time record of 11.5" on May 5, 1917, remains untouched.

Just as impressive as the snow itself was the juxtaposition of winter weather and severe weather within the same state. Snow and tornadoes were being observed at the same time (in different locations) in Colorado on Saturday afternoon and South Dakota on Sunday morning. I experienced this contrast first-hand while driving back to the Boulder, CO, area on Saturday from my visit to the NOAA Hazardous Weather Testbed in Norman, OK. Our route through far eastern Colorado took us just east of a long-lived, northward-moving supercell that produced several highly visible tornadoes north of the town of Eads. I once spent an entire summer storm-chasing in Colorado for NOAA as part of a validation program for the prototype NEXRAD radar, but this was the most spectacular sequence of tornadoes I've ever seen here. The last part of the drive was the most stressful, as howling winds and heavy snow struck the Denver area just before we arrived. This was the first time I've gone from a tornado warning to a winter storm warning in the space of six hours!

Bob Henson (severe); Jeff Masters (tropical)


Figure 5. Another tornado from the supercell that produced the twister in Figure 4 above, photographed from near Cheyenne Wells, CO, on Saturday, May 9. Image credit: wunderphotographer adkinsadam1.


Figure 6. A snow-covered scene from the Denver suburb of Arvada on Sunday, May 10. Image credit: wunderphotographer NapalmSnowcone.

Tornado tropical cyclone

Tropical Storms, Tornadoes, a Cat 5 Typhoon, and a Blizzard, Oh My!

By: Jeff Masters and Bob Henson , 3:59 PM GMT on May 10, 2015

Meteorological Mother's Day Madness is at hand. The strongest tropical storm ever to make landfall in the U.S. so early in the year hit South Carolina this morning, at the same time as a Category 5 super typhoon hit the Philippines, a raging blizzard pounded South Dakota, and the U.S. Plains cleaned up from an onslaught of 45 tornadoes the previous day. Let's start today's post with Tropical Storm Ana, which made landfall at 6:15 am EDT on the coast of South Carolina just south of the North Carolina border. At landfall, Ana had top winds of 45 mph, making the the strongest tropical storm ever recorded to hit the U.S. so early in the year. Ana brought peak winds of 47 mph, gusting to 58 mph, at 9:50 pm EDT Saturday to the Frying Pan Shoals Buoy 41013, located off the coast of North Carolina. North Myrtle Beach, SC recorded a gust of 49.5 mph at 12:53 am this morning. Ana is expected to bring 2 - 4" of rain to portions of coastal South Carolina and North Carolina though Monday.


Figure 1. MODIS satellite image of Tropical Storm Ana taken near 3 pm EDT May 9, 2015. At the time, Ana had top winds of 60 mph. Image credit: NASA Worldview.


Figure 2. Radar image of Ana taken at 6:15 am EDT Sunday May 10, 2015, from Wilmington, North Carolina. Ana was making landfall near Myrtle Beach, South Carolina.

How unusual is a May landfall?
Named storms that make landfall in the U.S. before the June 1 start of hurricane season are rare; between 1900 - 2014 there were only nine such landfalls, an average of one every thirteen years. The only U.S. landfall on record that occurred earlier than Ana's was on February 3, 1952, when the Groundhog Day tropical storm, packing winds of 40 mph, struck near Cape Sable, Florida. Eight U.S. landfalls have occurred in May, with Beryl the strongest of the group in recent years, packing 65-mph winds during its landfall near Jacksonville Beach, Florida, on May 28, 2012 (note, though, that a 2013 paper in the Bulletin of the American Meteorological Society documented a hurricane landfall on the Florida Panhandle on May 28, 1863, with estimated peak winds of 105 mph.)

Here is the full list of peak winds observed on U.S. shores from named systems before June 1 since 1900, including both landfalling and offshore systems, as compiled by tropical meteorologist Andrew Hagen (StormGeo) using the HURDAT database. These are ranked in descending order of observed winds and listed with the state(s) where the winds were recorded.

5/28/2012 – 55 kt – FL (Beryl)
5/29/1908 – 55 kt – NC
5/27/1972 – 50 kt – GA (Alpha)
5/10/2015 – 40 kt – NC (Ana)
5/23/1976 – 40 kt – FL/GA (unnamed)
5/30/1959 – 40 kt – LA (Arlene)
2/3/1952 – 35 kt – FL (Groundhog Day Storm)
5/10/2007 – 35 kt – GA (Andrea)
5/14/1916 – 35 kt – FL
5/17/1951 – 35 kt – FL (Able)


Figure 3. Super Typhoon Noul as seen by the new Japanese Himawari-8 satellite at 00:50 UTC May 10, 2015. At the time, Noul was a Category 5 storm with 160 mph winds. The University of Wisconsin/CIMSS has put together a remarkable hi-res satellite animation of the storm from the Himawari-8 satellite (which is still in check-out mode.) This is the most advanced geostationary satellite ever launched, and has images available every ten minutes with a resolution four times higher than the U.S. GOES satellites. The Himawari-8 imager is a clone of the imager expected to be launched on the U.S. GOES-R satellite in March 2016, which is likely to be initially positioned to view the Western Atlantic and Eastern Pacific. Something to look forward to!

Category 5 Super Typhoon Noul hits the Philippines
Category 5 Super Typhoon Noul (called Dodong in the Philippines’ naming system) made landfall near 5 am EDT Sunday, May 10, on the extreme northeastern tip of the Philippines' Luzon Island at Pananapan Pt., Sta. Ana, Cagayan. Satellite loops on Sunday morning showed that Noul's encounter with land significantly weakened the storm, with the eye disappearing from view and the storm's heavy thunderstorm activity noticeably deteriorating. Storm chaser James Reynolds has commentary and video of the landfall at his Twitter feed. Thankfully, the northeastern portion of the Philippines is sparsely populated, so Noul is not likely to be a major disaster for the Philippines. Noul is caught in a trough of low pressure that will swing the storm to the north and northeast, out to sea, skirting the coast of Japan.

An unusually early and violent typhoon season
May 10 is exceptionally early to be getting a Category 5 storm in the Northwest Pacific, let alone one landfalling at that strength! At its peak on Sunday morning, the Japan Meteorological Agency put Noul's central pressure at 915 mb. This makes Noul is the third strongest typhoon on record for so early in the year. The only stronger ones were Super Typhoon Amy of 1971, which deepened to 890 mb on May 2, and Super Typhoon Iris of 1951, which hit 909 mb on May 3.

Super Typhoon Noul is already the second Category Five storm this year in the Northwest Pacific, and the fourth overall on Earth in 2015. This is an unusually large number of these high-end tropical cyclones for so early in the year; Earth averaged just 4.6 Category 5 storms per year between 1990 - 2014, with the majority of these storms occurring during the August - November peak of the Northern Hemisphere's tropical cyclone season; 59% of all Cat 5s occur in the Northwest Pacific. Noul's formation date of May 3 was the 2nd earliest appearance of the season's 6th named storm on record for the Northwest Pacific. The Japan Meteorological Agency (JMA) just began issuing advisories for Tropical Storm Dolphin in Micronesia on May 9, breaking the record set on May 19, 1971 for the earliest formation of the Northwest Pacific's seventh named storm of the year, according to statistics of the Japan Meteorological Agency's database from 1951 - 2015 maintained by Digital Typhoon. The early and violent start to 2015 typhoon season is due, in part, to exceptionally warm ocean temperatures in the typhoon breeding region between 5 - 10°N near the Date Line. These temperatures have been over 2°C (3.6°F) above average in recent months, due to a strengthening El Niño event.

The other three Category 5 storms of 2015 were Super Typhoon Maysak of March (160 mph winds), one of only three Category 5 typhoons ever observed in the Northwest Pacific prior to April; Tropical Cyclone Pam (165 mph winds), which devastated Vanuatu in mid-March, and Tropical Cyclone Eunice (160 mph winds), which affected ocean areas in the South Indian Ocean. Reliable satellite records of Southern Hemisphere tropical cyclones extend back to the early 1990s, so we only have about a 25-year period of good records for global Category 5 tropical cyclones.


Figure 4. Radar image of the supercell thunderstorm that hit Cisco, Texas on Saturday night, killing one person and injuring three.

Multiday severe outbreak shifts to eastern Plains
Storm-hammered residents of Oklahoma and Texas will get a chance to catch their breath early this week, as the intense weather plaguing the area since Wednesday finally moves eastward. The next significant episode is likely to arrive in the Southern Plains around next weekend, possibly preceded by yet another batch of heavy-raining storms later this week. In the meantime, severe storms will cover an expanded region on Sunday, with a slight risk extending from southern Minnesota to Illinois to central Texas, as a strong upper low accelerates eastward from Colorado. In its 1:00 am CDT outlook, the NOAA Storm Prediction Center called for an enhanced risk on Sunday in two areas, one centered on western Iowa (near the surface low and warm front) and another on northeast Texas (where the trailing end of the cold front will encounter especially rich low-level moisture). The highest risk of tornadoes will be in these two areas, although a widespread outbreak of strong tornadoes is not expected. Sunday's severe weather got off to an early start, though, with three tornado watches already in effect by 11:15 am CDT and law enforcement reporting damage from a tornado that struck near Delmar, SD, around 10:30 am CDT.

Saturday's severe weather was also split into two main focal points. Toward the south, a fearsome supercell churned just south of Interstate 20 from east of Abilene, TX, to the southwest exurbs of Dallas-Fort Worth. At least one person was killed by a tornado near the town of Cisco (see Figure 1) that appears to have produced significant damage. Although the storm later produced another tornado near Stephenville, it fortunately failed to live up to the potential implied by a very strong and persistent rotation signature on Doppler radar. Another supercell about 100 miles to the north generated twisters near Burkburnett, TX, and Randlett, OK, and a brief tornado was also reported near Elk City, OK. The other center of action for tornadoes was across eastern Colorado and western Kansas, where several storms were prolific twister-producers. Strong winds and cold air associated with the upper low more than compensated for cool surface temperatures (mainly in the 60s F), producing a very favorable setup for multiple high-plains tornadoes. An isolated supercell moving north from Eads, CO, produced at least six tornadoes across largely unpopulated terrain, and more than a dozen tornado reports came in from a late-evening storm that crossed I-70 in far northwest Kansas. Several more tornadoes were spun out by another cell just to the east (see Figure 5 below).


Figure 5. One of several tornadoes that emerged from a supercell near Collyer, KS. Image credit: wunderphotographer mhaase7.

Just a couple hours' drive northwest of the Colorado twisters, the scene was decidedly more wintry by evening. A burst of heavy snow swept across the Denver metro area around sunset, causing numerous accidents as temperatures fell into the upper 20s. The snow was preceded by heavy rain that put much of Colorado in flash flood watches and warnings. At one point, Colorado Springs was under five different types of NWS alerts, as noted by the Denver Post. The snow extended into Colorado's northern mountains, western Nebraska, eastern Wyoming, and north to the Black Hills of South Dakota, where blizzard warnings were in effect on Saturday night. Snowfall in May isn't all that uncommon in these parts--Denver averages 1.7 inches for the month--but this weekend's onslaught in Rapid City was another in a string of noteworthy snows falling either very early or very late in the season. These include the all-time multiday record storm of April 8-10, 2013 (28.2"); the devastating snowstorm of October 3-5, 2013 (23.1"); and the city's earliest snowfall on record (1.6" on September 11, 2014).

We're again covering today's severe weather on the experimental WU live blog.

Jeff Masters (tropical) and Bob Henson (severe)


Figure 6. The late-evening light of May illuminates snow falling on Saturday in Minatare, Nebraska, just southeast of Scottsbluff. Image credit: wunderphotographer JdyJdyJdy.

Hurricane Severe Weather Winter Weather Tornado

Ana a Tropical Storm; Complicated Severe Threat for U.S. Plains

By: Jeff Masters and Bob Henson , 4:22 PM GMT on May 09, 2015

Tropical storm conditions are expected beginning Saturday evening along much of the South Carolina and North Carolina coasts as slow-moving Tropical Storm Ana plods to the northwest at 3 mph. Ana gained enough heavy thunderstorms near its center on Saturday morning to be classified as a tropical storm, but long-range radar out of Wilmington, North Carolina and satellite loops on Saturday afternoon showed that these thunderstorms were of only modest intensity and areal coverage, and were only slowly increasing. Strong upper level winds out of the north were creating an increasing amount of wind shear over Ana on Saturday morning, and the shear had reached a high 20 knots by 8 am EDT. This shear was driving dry air into the northern side of Ana, and keeping heavy thunderstorm activity limited on that side. Ocean temperatures were near 25°C (77°F), which is just at the limit of where a tropical storm can sustain itself. The Frying Pan Shoals Buoy 41013, located off the coast of North Carolina about 40 miles northeast of the center of Ana, recorded sustained winds of 38 mph, gusting to 40 mph, at 9:50 am EDT. Wunderblogger Steve Gregory pointed out to me this morning that significant wave heights in Ana have not been very impressive over the past few days, less than ten feet, suggesting that the advertised peak winds of Ana of 60 mph were either too high or were affecting a very small portion of the ocean.


Figure 1. Subtropical Storm Ana as seen by the International Space Station on Thursday afternoon, May 8, 2015. At the time, Ana had top winds of 45 mph. Image credit: Scott Kelly.


Figure 2. Radar image of Ana taken at 12:03 pm EDT Saturday May 9, 2015, from Wilmington, North Carolina. A modest rainband was affecting the coast of North Carolina, but the main activity was still out to sea.

Forecast for Ana
The 8 am EDT Saturday run of the SHIPS model predicted that wind shear will remain a moderately high 20 knots until the storm makes landfall early Sunday morning, and ocean temperatures beneath the core of the storm will cool to 22°C (72°F) by Sunday morning. These two effects should act to weaken Ana before landfall, and the storm should have sustained winds no stronger then 50 mph at landfall. The 0Z Saturday morning runs of our two top models for predicting tropical cyclone tracks, the European and GFS models, both showed the system making landfall on Sunday between 1 am - 5 am EDT near the North Carolina/South Carolina border. With winds of 50 mph, rainfall of 1 - 3 inches, and a storm surge of 1 - 2 feet likely the worst impacts that Ana can deliver, the main danger of the storm may be rip currents that can drown unwary swimmers.


Figure 3. Typhoon Noul as seen by the International Space Station on Thursday, May 8, 2015. At the time, Noul was a Category 3 storm with top winds of 115 mph. Image credit: Terry Virts.

Category 4 Typhoon Noul headed for the northern Philippines
Category 4 Typhoon Noul (called Dodong in the Philippines’ naming system) continued a slow intensification process on Saturday, with winds estimated at 135 mph as of 8 am EDT. Satellite loops on Saturday afternoon showed that Noul was a medium-sized storm with a 14-mile diameter eye and a modest area of heavy thunderstorms. Landfall of the center is expected to occur in northeastern Luzon Island near 2 am EDT Sunday (06 UTC), according to the 0Z Saturday run of the GFS model. Noul is encountering a trough of low pressure as it approaches the Philippines, which is inducing a more northwesterly track. This may cause the storm's eye to miss the Philippines, as predicted by the 0Z Saturday run of the European model. In either case, Luzon will be on the weaker left front side of Noul, which will make heavy rains the main threat from the storm. I do not expect this to be a major disaster for the Philippines; loss of life should be limited and damage in the tens of millions of dollars even if Noul makes a grazing landfall as a Category 4 storm. With wind shear still a light 5 - 10 knots and warm ocean waters ahead of it, Noul should be able to hold on to Category 4 strength until just a few hours before landfall.

Widespread rains complicate severe potential for this weekend
Round after round of heavy rain has triggered flooding in communities across the southern Plains while also denting the atmosphere’s potential to product tornadic supercells. Once again on Friday, a large convective complex developed by midafternoon before more classic, discrete supercells had a chance to form. The complex spun out several brief tornadoes along its southern fringes while dumping several inches of rain from north Texas across much of Oklahoma. The sense of deja vu was especially strong in northern parts of Norman, OK, where the same areas struck by flooding and an EF1 tornado on Wednesday experienced hail larger than golf balls, winds to 73 mph, and 1” - 2” of rain on Friday, knocking out windows and knocking down power lines. At the other end of the storm complex, softball-sized hail was reported near Slaton, TX (southeast of Lubbock). Rainfall totals for Tuesday through Saturday morning ranged from 5” to more than 10” across parts of the southern half of Oklahoma and adjacent north Texas.


Figure 4. Pastures near Chico, TX, northwest of the Dallas-Fort Worth area, were inundated by Friday, May 9, after multiple days of heavy rain. Image credit: wunduerphotographer Madermade.

A challenging severe weather forecast
The upper low that's driven the persistent storminess is finally on the move, pushing into the western Plains. Upper winds will be much stronger later today, raising the potential for a few significant tornadoes if the atmosphere manages to recover, but that’s a big “if.” With little in the way of a warm layer at midlevels to inhibit thunderstorm development until peak afternoon heating, a sprawling area of heavy showers and thunderstorms developed overnight from the Texas Panhandle into western Kansas and northeast Colorado. These storms are cooling and dampening a wide part of the region that would otherwise be at risk of late-day supercells. In its 1:00 am CDT Saturday outlook, NOAA's Storm Prediction Center called it "another messy/complex forecast." The 1:00 am outlook had placed southwest Kansas under a moderate risk of severe weather, but the 8:00 am CDT update eliminated that area, leaving an enhanced risk from eastern Colorado to north Texas. Behind this area, some afternoon warming may still occur, with redevelopment possible along a dry line/cold front. Further north, instability will be marginal but wind shear very favorable for relatively small rotating storms near the surface low and warm front pushing north across western Kansas and eastern Colorado. Toward the south, another large complex of storms may emerge by Saturday night, again heading toward the hard-hit region from the Dallas-Fort Worth area to the Red River. Heavy precipitation is also walloping northeast Colorado, where flash flooding is considered likely on Saturday afternoon and evening by the NOAA Weather Prediction Center. Heavy snow will envelop the mountains and higher foothills of northern Colorado and southeast Wyoming, as well as the Black Hills of South Dakota, with winter storm warnings covering much of this region. Even the Denver-Boulder area could see several wet inches of snow overnight Saturday night.

Severe weather on Sunday will expand into a larger swath of the central United States, from Iowa to Texas, although the areas of peak intensity and coverage will again depend largely on how storms evolve through Saturday night and Sunday morning. Any substantial tornado threat would most likely be centered in Iowa. Severe storms will again be widespread on Monday toward the Mississippi Valley, but that should be the last day of this prolonged, complex episode, as the upper low lifts out and wind shear relaxes.


Figure 5. More than a foot of snow is possible across the higher terrain of the central Rockies and adjacent Black Hills from Saturday to Sunday morning. Image credit: NOAA Weather Prediction Center..

Wunderground will be running a live blog Saturday afternoon from our main page to track the severe weather.

Jeff Masters (tropical) and Bob Henson (severe)

Hurricane Severe Weather

Subtropical Storm Ana More Organized; Philippines' Cat 3 Noul Intensifying

By: Jeff Masters , 6:44 PM GMT on May 08, 2015

Tropical storm warnings are flying along much of the South Carolina and North Carolina coasts as a nearly stationary Subtropical Storm Ana spins away about 180 miles south of Myrtle Beach, South Carolina. Ana gained enough organization and heavy thunderstorms last night to be classified as a subtropical storm; its formation date of May 7 is the earliest appearance of a named storm in the Atlantic since a previous incarnation of Subtropical Storm Ana was recognized on April 20, 2003.

Long-range radar out of Wilmington, North Carolina Friday afternoon showed only a small amount of heavy rains associated with the storm were reaching the coast, and satellite loops showed that the main area of heavy thunderstorms were on the southeast side of the storm, farthest from the coast, and about 100 miles from the center of circulation. This is a typical appearance for a subtropical cyclone. As explained in wunderground's subtropical storm tutorial, a subtropical cyclone has characteristics of both tropical and extratropical cyclones. The difference between a subtropical storm and a tropical storm is not that important as far as the winds they can generate, but tropical storms generate more rain, and tropical storms have the potential to rapidly intensify into hurricanes, while subtropical storms do not. The area of heavy thunderstorms was expanding in areal coverage and a tight circulation center fully exposed to view was apparent on Friday afternoon, indicating that Ana was growing more organized; Ana will likely be a tropical storm and not a subtropical storm by Friday night. Wind shear over Ana was a light 5 - 10 knots, which was aiding development. Ocean temperatures were near 25°C (77°F), which is just at the limit of where a tropical storm can form. Cold air aloft associated with an upper level low was increasing the instability of the atmosphere, though, allowing Ana's heavy thunderstorms to grow more than usual for ocean temperatures of 25°C. This upper-level low was also bringing dry air into Ana's core, though, which has been slowing the storm's development.


Figure 1. Latest satellite image of Ana.


Figure 2. Radar image of Ana taken at 2:16 pm EDT Friday May 8, 2015, from the Long-range radar out of Wilmington, North Carolina. A few modest rainbands were affecting the coasts of South Carolina and North Carolina.

Forecast for Ana
The 8 am EDT Friday run of the SHIPS model predicted that wind shear over Ana would be in the light to moderate range, 5 - 20 knots, until the storm makes landfall on Sunday. Steering currents are weak over the waters off the Southeast U.S. coast, so expect a slow motion for the storm. The 12Z Friday morning runs of our two top models for predicting tropical cyclone tracks, the European and GFS models, both showed the system making landfall on Sunday between 8 am - 5 pm EDT near the North Carolina/South Carolina border. Given the relatively marginal environmental conditions for intensification, the cooler waters Ana will encounter as it nears the coast, and the fact the storm has only about two days until landfall, I expect that the worst Ana will be able to do is intensify to a 65-mph tropical storm that brings 4 - 6" of rain to the coasts of South Carolina and North Carolina on Saturday and Sunday. My expectation is that the storm will have 55 mph winds at landfall. In their 11 am EDT Friday forecast, NHC called for 50 mph winds at landfall.

Wunderground member Levi Cowan has been putting together some nice video discussions of Ana; you can check out his 11 am EDT Friday version here.


Figure 3. MODIS satellite image of Typhoon Noul taken from NASA's Aqua satellite at 04:55 UTC May 8, 2015. At the time, Noul was a Category 3 storm with 115 mph winds. Image credit: NASA.

Category 3 Typhoon Noul headed for the northern Philippines
Category 3 Typhoon Noul (called Dodong in the Philippines’ naming system) continued a slow intensification process on Friday morning, with winds estimated at 120 mph winds as of 8 am EDT Friday. WIth plenty of deep, water water before it and wind shear a light 5 - 10 knots, continued intensification into a Category 4 storm appears likely as the storm heads towards the northern Philippines. Satellite loops on Friday afternoon showed that Noul was a medium-sized storm with a 16-mile diameter eye and a modest area of heavy thunderstorms. The storm didn't have much in the way of outer spiral bands, and had taken on the shape of a large donut. This sort of configuration may qualify Noul as an uncommon class of tropical cyclone called an "annular" storm. Annular storms are more able to resist weakening, and I expect Noul will remain at least a Category 3 storm with 115 mph winds within six hours of landfall, despite the increasing levels of wind shear and weakening effects of land interaction that will be occurring. The outer reaches of Noul are already affecting the Philippines, and landfall of the center is expected to occur in northeastern Luzon Island near 11 pm EDT Saturday (03 UTC Sunday), according to the 12Z Friday run of the GFS model. Noul will encounter a trough of low pressure as it approaches the Philippines, which is expected to induce a more northwesterly track. This may cause the storm's eye to miss or barely graze the Philippines, as predicted by the 12Z Friday run of the European model. In either case, Luzon will be on the weaker left front side of Noul, which will make heavy rains the main threat from the storm. The 06Z Friday run of the GFDL model predicted that heavy rains in excess of 8 inches would be confined to a very small portion of the northeast coast of Luzon. I do not expect this to be a major disaster for the Philippines; loss of life should be limited and damage in the tens of millions of dollars if Noul makes a grazing landfall as a Category 3 or 4 storm.

Noul will be the second tropical cyclone to affect the Philippines so far in 2015. The first was Tropical Storm Maysak, which hit the Philippines exceptionally early in the season--during Easter weekend, April 4 - 5. Fortunately, Maysak was weakening rapidly as it made landfall, and no deaths or significant damage were reported (though four people were injured after huge waves generated by Maysak hit them while they were taking selfies along the shoreline of Dipaculao town in Aurora province on April 4.)

Record early activity for the 2015 typhoon season
On the heels of Noul comes Tropical Storm Seven, which is organizing to the east of Noul over Micronesia. TS 7 should pass within 300 miles of Guam next week as an intensifying typhoon, and may pose a long-range threat to the Philippines. Satellite loops show that TS 7 is not yet well-organized, but it does have a large area of heavy thunderstorms. The Japan Meteorological Agency (JMA) was still classifying TS 7 as a tropical depression on Friday afternoon, but when JMA gives TS 7 a name, it will break the record set on May 19, 1971 for the earliest formation of the Northwest Pacific's seventh named storm of the year, according to statistics of the Japan Meteorological Agency's database from 1951 - 2015 maintained by Digital Typhoon. The early start to 2015 typhoon season is due, in part, to exceptionally warm ocean temperatures in the typhoon breeding region between 5 - 10°N near the Date Line. These temperatures have been over 2°C (3.6°F) above average in recent months, due to a strengthening El Niño event.

All-time May heat record for Europe falls--in the first week of May!
Extreme heat roasted Italy this week, where the city of Catenanuova hit 41.9°C (107.4°F) on Wednesday, May 6. According to weather records researcher Maximiliano Herrera, this is the hottest May temperature ever recorded in Europe. The previous record was 41.7°C (107.1°F), set on May 17, 2006 in Andujar, Spain. The new European May heat record comes just one day after the earliest 40°C temperature ever recorded in Europe--a 40.4°C (104.7°F) reading at San Priamo in southeastern Sardinia, according to Herrera's stats.

A major severe weather outbreak is expected across the Central U.S. on both Friday and Saturday afternoon, as discussed by Bob Henson in this morning's blog post. Check out our Severe Weather Page to follow the storms, and our live blog on Saturday.

Jeff Masters

Hurricane

Volatile Setup for Severe Weather in Southern Plains on Friday, Saturday

By: Bob Henson , 1:49 PM GMT on May 08, 2015

The ingredients are in place for a major multi-day outbreak of severe weather from Friday into the weekend, including the possibility of strong, long-track tornadoes. Early on Friday, NOAA’s Storm Prediction Center placed part of southwest Oklahoma under a moderate risk of severe weather for Friday, with a moderate-risk area covering a broader swath from western KS to southwest OK on Saturday. I would not be surprised to see a localized upgrade to high risk somewhere in this swath on Saturday.


Figure 1. Convective outlooks for Friday and Saturday, May 8 and 9, issued early Friday morning by NOAA’s Storm Prediction Center. Image credit: NOAA/SPC.


The tornado potential for Friday will be heavily influenced by an outflow boundary that formed late Thursday, separating rain-cooled air over Oklahoma and the Texas Panhandle (temperatures and dew points early Friday in the 50s to low 60s F) and very warm, humid air just to the south (temperatures and dew points in the low 70s). This boundary will attempt to push back north as a warm front on Friday and should be positioned somewhere near the Red River area of Oklahoma/Texas by late afternoon. Near this boundary, surface air will converge and surface winds will have a strong easterly component. With southwesterly winds expected just a few thousand feet above the surface, this will enhance the vertical wind shear that produces rotation in supercell thunderstorms. The potential for strong tornadoes is greatest when strong vertical shear is present in the lowest kilometer of the atmosphere, and when high temperatures and moisture values are also present in that lowest kilometer. Together, these tend to produce thunderstorms with low cloud bases and intense rotation near those bases. Any supercell that roots itself near the warm front and moves east or northeast along it will be a particular threat for producing tornadoes, although tornadic supercells are possible throughout the moderate-risk area. Flash flooding could be a serious threat as well, especially across parts of central Oklahoma and north-central Texas that have seen 6” to 10” of rain over the last several days (see Figure 3 below). One important caveat: several of the 00Z Friday model runs, as well as several recent runs of the hourly-updated HRRR and RAP mesoscale models, suggest that a large storm complex will form early in the day across west Texas and move across western Oklahoma, as was the case on Thursday. Storms were beginning to form west of Lubbock and south of Abilene at 8:30 am CDT, supporting this scenario. Should this activity increase, it would act to reinforce the outflow boundary and could markedly suppress the risk of supercells and tornadoes across parts of the current moderate and enhanced risk areas, though supercells could still form in pockets where the air mass is less affected. This situation may become quite geographically complex through the day, adding to the challenge for forecasters.

A sprawling severe threat for Saturday
Forecast models have agreed for the last several days that the stubborn upper low over the Southwest will finally make its move onto the Plains on Saturday. Instability may be a bit less than on Friday, but still more than adequate for high-end supercells. Moreover, howling upper-level winds and strong dynamics associated with the low will lead to very high vertical wind shear and dangerous, fast-moving storms over a broad area along and in front of a sharpening surface low in eastern Colorado and a dry line extending south across western Oklahoma into north Texas. Many cells could be moving at 40 - 50 mph. Widespread severe weather is almost a certainty on Saturday, and the risk of strong, long-track tornadoes may also cover an unusually large area (see Figure 2). Faster storm motion will help reduce the flash-flood risk somewhat, although storms rooted near the surface low will be slower-moving. Once again, if morning thunderstorms turn out to be quite extensive, the risk of significant severe weather will be tamped down.


Figure 2. Surface winds (flags) and significant tornado parameter (STP, in colored areas) at 7:00 p.m. CDT Saturday, May 9, as projected by the 0000 GMT Friday run of the NAM model. The STP is based on several measures of wind shear, instability, and cloud-base height. STP values of 8 to 11, as predicted here for western Oklahoma, are near the top end of those seen in major outbreaks, such as the Super Outbreak of April 2011 (see an example from that day). Image credit: College of DuPage.


Sunday could bring yet another round of severe weather, with the focus this time pushing toward eastern Kansas and Nebraska, Missouri, and Iowa. The risk on Sunday will be highly contingent on the amount of rain-cooled air produced by storms late Saturday toward the south. If the air mass can recover by Sunday afternoon, wind shear and instability again look favorable for potentially tornadic supercells, especially along the warm front that should be across Iowa by then.

Weather Underground is planning to cover the severe weather outbreak on Saturday in our experimental live blog, which you can access through a banner at the top of the home page. We’ve also activated a live blog on Subtropical Storm Ana. In addition, Jeff Masters and I will be posting updates at this blog on both events during the weekend. Jeff will have a complete update later today on Subtropical Storm Ana, which continues to gradually develop off the Southeast coast. Ana was being investigated by a hurricane-hunter aircraft on Friday morning.

Bob Henson


Figure 3. Flooding on Thursday night, May 7, in downtown Gainesville, TX, just south of the Oklahoma-Texas border. Photo credit: @TheMaverick21, used with permission.

Severe Weather Tornado

Subtropical Storm Ana Forms Near South Carolina

By: Jeff Masters and Bob Henson , 3:03 AM GMT on May 08, 2015

The area of disturbed weather (90L) off the Southeast U.S. coast finally gained a well-defined circulation center and enough heavy thunderstorms near its core to be designated Subtropical Storm Ana at 11 pm EDT Thursday. Ana's formation date of May 7 is the earliest appearance of a named storm in the Atlantic since a previous incarnation of Subtropical Storm Ana was recognized on April 20, 2003.

Long-range radar out of Charleston Thursday evening showed only a modest amount of heavy rains associated with the storm, and satellite loops showed a large circulation with most of the heavy thunderstorms more than 100 miles away from the center. This is a typical appearance for a subtropical cyclone. As explained in wunderground's subtropical storm tutorial, a subtropical cyclone has characteristics of both tropical and extratropical cyclones. The difference between a subtropical storm and a tropical storm is not that important as far as the winds they can generate, but tropical storms generate more rain, and tropical storms have the potential to rapidly intensify into hurricanes, while subtropical storms do not. Wind shear over Ana was a moderate 20 knots late Thursday night, and water vapor satellite loops showed a large area of dry air to the west over the Southeast United States. This dry air is retarding development, thanks to strong upper-level winds out of the west driving the dry air into the storm's core. Ocean temperatures were near 25 - 26°C (77 - 79°F), which is just at the limit of where a tropical storm can form. Given these relatively cool ocean temperatures and the fact Ana will need at least a day to transition to a fully tropical storm, I expect that the worst Ana will be able to do is intensify to a 65-mph tropical storm that brings 4 - 6" of rain to the coasts of South Carolina and North Carolina on Saturday and Sunday. The 11 pm EDT Thursday forecast from NHC, which calls for a 50-mph storm bringing 2 - 4" of rain to the coast, is more likely.


Figure 1. Latest satellite image of Ana.



Figure 2. Sea Surface Temperatures (SSTs) from the AVHRR satellite instrument clearly show the warm Gulf Stream current off the Southeast U.S. coast for the 6-day period ending on May 2, 2015, before the arrival of 90L (AKA Ana). The storm took advantage of the warm waters of the Gulf Stream to aid its intensification into Subtropical Storm Ana. Image credit: Johns Hopkins University Applied Physics Laboratory.

How unusual would a landfall be?
It’s not so rare to get a tropical or subtropical cyclone developing somewhere in the Atlantic basin before the official June 1 start of the hurricane season. Since 1851, a total of 39 “pre-season” systems have developed across 32 separate years. We’ve averaged about one pre-season tropical or subtropical system in the Atlantic every two to three years since the modern satellite era began in 1960. Interestingly, among these early-starting seasons, 38% ended up producing more storms than a typical season, while 62% produced near- or below-average totals. Pre-season named storms may be getting more common. In 2008, Dr. Jim Kossin of the University of Wisconsin published the paper, "Is the North Atlantic hurricane season getting longer?" in the journal Geophysical Research Letters. He concluded that there is an "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".

Very few of the tropical/subtropical systems that develop prior to June 1 make landfall. The earliest U.S. landfall on record occurred on February 3, 1952, when the Groundhog Day tropical storm struck near Cape Sable, Florida. Several U.S. landfalls have occurred in May, with Beryl the strongest of the group in recent years, packing 65-mph winds during its landfall near Jacksonville Beach, Florida, on May 28, 2012. However, a recent paper in the Bulletin of the American Meteorological Society documented a hurricane landfall on the Florida Panhandle on May 28, 1863, with estimated peak winds of 105 mph (see Figure 3).


Figure 3. Track of the May 28, 1863 hurricane, which struck northwest Florida as a Category 2 storm with 105 mph winds, killing at least 72 people. This was the only hurricane on record to hit the U.S. in May, and stuck nearly two weeks earlier than the next earliest U.S. landfalling hurricane, Hurricane Alma of June 9, 1966. (Tropical Storm Beryl of May 28, 2012 came close to being a May hurricane, bringing 65 mph winds to the coast near Jacksonville Beach, Florida.) According to a 2013 paper by hurricane historians Mike Chenoweth and C. J. Mock in the Bulletin of the American Meteorological Society, "Among the most unusual and unexpected hurricanes in United States history is the only hurricane to make landfall in the month of May. This recently re-discovered storm that struck northwest Florida on 28 May 1863 created a natural disaster in the area that became lost to history because it was embedded in a much larger and important manmade event, in this case the U.S. Civil War. We document the arrival of this storm both historically and meteorologically and anachronistically name it Hurricane “Amanda” in honor of the Union ship driven ashore by the hurricane. The hurricane revealed deficiencies and strengths in combat readiness by both sides. Meteorologically, the storm nearly achieved major hurricane status at landfall and its absence from modern data bases of tropical cyclone activity is a useful reminder to users of important gaps in our knowledge of tropical cyclones even in the best-sampled storm basins." Image credit: Mike Chenoweth and the Bulletin of the American Meteorological Society.

Here is the full list of peak winds observed on U.S. shores from named systems in May since 1900, including both landfalling and offshore systems, as compiled by tropical meteorologist Andrew Hagen (StormGeo) using the HURDAT database. These are ranked in descending order of observed winds and listed with the state(s) where the winds were recorded. If the current low were to develop into Ana and bring winds of at least 45 mph (40 knots) onshore, it will be the strongest U.S. impact by a tropical/subtropical storm on record for so early in the season.

5/28/2012 – 55 kt – FL (Beryl)
5/29/1908 – 55 kt – NC
5/27/1972 – 50 kt – GA (Alpha)
5/23/1976 – 40 kt – FL/GA (unnamed)
5/30/1959 – 40 kt – LA (Arlene)
5/10/2007 – 35 kt – GA (Andrea)
5/14/1916 – 35 kt – FL
5/17/1951 – 35 kt – FL (Able)

Only five tropical cyclones since 1851 have reached hurricane strength prior to June 1. The only one that exceeded Category 1 on the Saffir-Simpson scale was Hurricane Able of 1951, which carried out a counter-clockwise loop off the Florida coast before swinging northeast just off the North Carolina coast. Able attained peak winds of 115 mph about 70 miles east of Cape Hatteras, NC, on May 22. As noted above, Able also produced 40-mph (35 knot) winds in Florida, though it didn’t make landfall there. Many thanks to Andrew Hagen (StormGeo), Michael Lowry (Weather Channel), and wunderground member TropicalAnalystwx13 for information used above.

Jeff Masters and Bob Henson

Hurricane

Hurricane Hunters Find 50 mph Winds in 90L; Cat 3 Noul Aims at Philippines

By: JeffMasters, 4:31 PM GMT on May 07, 2015

An Air Force hurricane hunter aircraft measured winds in excess of tropical storm force on Thursday morning inside Invest 90L, the area of disturbed weather off the Southeast U.S. coast. However, the aircraft found no well-defined center, and 90L lacked enough heavy thunderstorm activity for NHC to name the system Subtropical Storm Ana as of 11 am EDT Thursday. The aircraft found a large area of 40 - 50 mph surface winds about 100 - 200 miles off the central and northeast coast of South Carolina. Winds of this strength were also measured by a 7:23 am EDT Thursday pass of the Rapidscat instrument on the International Space Station. (Note that in their In their 2 pm EDT Thursday Tropical Weather Outlook, NHC judged that rain contamination of the wind data meant these winds were more likely 40 - 45 mph.) As of noon EDT Thursday, rainfall amounts of about one inch were estimated by radar to have fallen along the South Carolina coast to the northeast of Charleston from 90L, but long-range radar out of Charleston showed little change in the intensity of organization of the rain bands of 90L. Satellite loops show that 90L has a large, ill-defined circulation, with a thick band of heavy thunderstorms extending to the northwest of the center over South Carolina, and almost no heavy thunderstorms near the center of circulation. This is a typical appearance for a subtropical cyclone, which characteristically have a large, cloud-free center of circulation, with heavy thunderstorm activity in a band removed at least 100 miles from the center. As explained in wunderground's subtropical storm tutorial, a subtropical cyclone has characteristics of both tropical and extratropical cyclones. The difference between a subtropical storm and a tropical storm is not that important as far as the winds they can generate, but tropical storms generate more rain, and tropical storms have the potential to rapidly intensify into hurricanes, while subtropical storms do not.

Wind shear over 90L had fallen to a moderate 20 knots late Thursday morning, and water vapor satellite loops showed a large area of dry air to the west over the Southeast United States. This dry air is retarding development, thanks to strong upper-level winds out of the west driving the dry air into the core of 90L. Ocean temperatures were near 25 - 26°C (77 - 79°F), which is just at the limit of where a tropical storm can form.


Figure 1. MODIS satellite image of 90L taken at approximately 11:30 am EDT May 7, 2015. Image credit: NASA Worldview.


Figure 2. Radar image of 90L taken at 12:02 pm EDT Thursday May 7, 2015, from the long-range radar out of Charleston.


Figure 3. The 7:23 am EDT (11:23 UTC) Thursday pass of the Rapidscat instrument on the International Space Station showed winds of up to 55 mph (50 knots, purple colors) off the coast of South Carolina from 90L. These winds may be weaker, perhaps 45 - 50 mph, due to contamination of the signal by heavy rains.

Forecast for 90L
The 8 am EDT run of the SHIPS model predicted that wind shear over 90L would fall to the light range, 5 -10 knots, on Friday, which should allow 90L to organize into a subtropical storm by Friday at the latest. Steering currents are weak over the waters off the Southeast U.S. coast, so expect a slow and erratic motion for 90L. The 00Z Thursday morning runs of our two top models for predicting tropical cyclones tracks, the European and GFS models, both showed the system making landfall on Sunday, in North Carolina or northeastern South Carolina. In their 2 pm EDT Thursday Tropical Weather Outlook, NHC gave the disturbance 2-day and 5-day odds of development of 80%, respectively. 90L will be capable of bringing winds of 55 - 65 mph to the coast on Saturday and Sunday, along with heavy rains of 4 - 6 inches. Steve Gregory provides additional perspective on 90L in his latest blog post.


Figure 4. MODIS satellite image of Typhoon Noul taken near 05 UTC May 7, 2015. At the time, Noul was a Category 2 storm with 105 mph winds. Image credit: NASA Worldview.

Dangerous Typhoon Noul headed for the northern Philippines
Typhoon Noul has undergone a period of rapid intensification, and had reached major typhoon status with 115 mph winds as of 8 am EDT Thursday. WIth plenty of deep, water water before it and wind shear a moderate 10 - 15 knots, continued intensification into a Category 4 storm appears likely as the storm heads towards the northern Philippines. Satellite loops show that Noul is a medium-sized storm with a tiny pinhole eye with a modest area of heavy thunderstorms. The outer bands of Noul will begin affection the Philippines on Friday, with landfall of the center expected to occur in northern Luzon Island near 6 pm EDT Saturday (22 UTC). Noul will encounter a trough of low pressure as it approaches the Philippines, which is expected to induce a more northwesterly track. This may cause the storm's eye to miss the Philippines, as predicted by the 00Z Thursday run of the European model. The northeast part of Luzon is mountainous, which could increase the risk for very heavy rains as Noul moves onshore or nearby, although a grazing landfall would put most of Luzon on the weaker western side of the circulation. Noul will be referred to as Dodong in the Philippines’ naming system.

Another tropical system, Tropical Depression Seven, is organizing to the east over Micronesia. TD 7 is expected to pass near Micronesia's Pohnpei Island this weekend as a Category 1 typhoon, and the storm will likely pass within 300 miles of Guam next week. Satellite loops show that TD 7 is not yet well-organized, but it does have a large area of heavy thunderstorms. TD 7 may pose a long-range threat to the Philippines or Japan, but it is too early to know.

An exceptionally busy early portion of typhoon season
Noul's formation date of May 3 marks the second earliest appearance on record for the Northwest Pacific's sixth named storm of the year, according to statistics of the Japan Meteorological Agency's database from 1951 - 2015 maintained by Digital Typhoon. The average is 1.8 storms before May 8. The record is held by 1971, when the sixth named storm of the year (Babe) formed on May 3, six hours earlier than Noul's formation time. If TD 7 develops into a named storm before May 19, which appears highly likely, it will break the record set on May 19, 1971 for the earliest formation of the Northwest Pacific's seventh named storm of the year. The early start to 2015 typhoon season is due, in part, to exceptionally warm ocean temperatures in the typhoon breeding region between 5 - 10°N near the Date Line. These temperatures have been over 2°C (3.6°F) above average, due to a strengthening El Niño event. Water temperatures are also unusually warm for this time of year across parts of the Northeast Pacific. This region has yet to see its first named storm of 2015, but the continued evolution of El Niño and the already-warm SSTs indicate that a busy hurricane season is quite possible in this area as well.

Noul will be the second tropical cyclone to affect the Philippines so far in 2015. The first was Tropical Storm Maysak, which hit the Philippines exceptionally early in the season--during Easter weekend, April 4 - 5. Fortunately, Maysak was weakening rapidly as it made landfall, and no deaths or significant damage were reported (though four people were injured after huge waves generated by Maysak hit them while they were taking selfies along the shoreline of Dipaculao town in Aurora province on April 4.)

All-time May heat record for Europe falls--in the first week of May!
Extreme heat roasted Italy this week, where the city of Catenanuova hit 41.9°C (107.4°F) on Wednesday, May 6. According to weather records researcher Maximiliano Herrera, this is the hottest May temperature ever recorded in Europe. The previous record was 41.7°C (107.1°F), set on May 17, 2006 in Andujar, Spain. The new European May heat record comes just one day after the earliest 40°C temperature ever recorded in Europe--a 40.4°C (104.7°F) reading at San Priamo in southeastern Sardinia, according to Herrera's stats.

Jeff Masters

Hurricane

Flooding, Tornado Threats Continue after Torrential Rains in Central Oklahoma

By: Bob Henson , 12:38 PM GMT on May 07, 2015

The Plains erupted on Wednesday with a batch of slow-moving tornadic supercells that morphed into prodigious rainmakers. More than four dozen tornado reports had been received by NOAA’s Storm Prediction Center by early Thursday. Pockets of wind/tornado damage were reported near Bridge Creek, OK, and across parts of the south metro area of Oklahoma City/Norman (see Figure 1). Two critical injuries and 10 other injuries were reported at a mobile home park and RV park in southeast Oklahoma City. The town of Rosedale, NE, also saw several homes damaged by a tornado. The more serious threat by late evening became flash flooding, especially across the Oklahoma City area, where a strip of heavy rainfall landed very close to a prior deluge the night before. The resulting 2-day totals (see Figure 2) exceeded 8” in spots. Oklahoma City’s total of 7.10” on Wednesday (including 6.03” in just three hours) was a record for any day in May and the third-highest all-time calendar-day total. Widespread flooding was reported across south parts of the metro area, and here in Norman, where I'm based this week, this was one of the heaviest downpours I've seen in a long time. Climate Central has a great roundup of statistics related to Oklahoma City's record rainfall. CoCoRaHS 24-hour rainfall reports as of 8:00 am CDT also show a large swath of torrential rain over southeast Nebraska, with several reports of between 8" and 11". Significant flooding occurred in the Lincoln, NE, area, with reports of water rescues under way on Thursday morning.


Figure 1. A hotel in northwest Norman, OK, received substantial exterior damage on Wednesday when a tornado passed just to its north. In 1991, this hotel (a Holiday Inn at the time) played host to the Third Fujita Tornado Symposium. Image credit: Trevor Hutson, used with permission.

The stubborn upper low parked over the Southwest isn’t expected to begin moving out until the weekend, so additional rounds of severe weather are possible over the next several days. The stronger upper-level winds expected to move over the Plains on Saturday will boost the potential for a significant outbreak of tornadic storms. In addition, many already-soaked areas will remain under the gun for the potential of even more heavy rain. While tornado threats often dominate severe-weather coverage, flash flooding can be the greater hazard. More people died from drowning than from tornadoes in the tragic May 31, 2013, storm that killed famed tornado chaser Tim Samaras and 20 other people. Atmospheric moisture will remain extremely rich across the south central U.S. until the upper low moves out, so the flooding risk will have to be taken very seriously.


Figure 2. Rainfall totals across Oklahoma for the 48 hours ending at midnight Wednesday night. Image credit: Oklahoma Mesonet.

Update on 90L
A hurricane hunter aircraft is in the air investigating Invest 90L, the nearly stationary area of disturbed weather off the coast of South Carolina. There will be a blog post by early afternoon detailing their findings. 90L is still on track to potentially become a subtropical depression by Friday, and bring heavy rain and high waves to South Carolina and North Carolina over the weekend. In their 8 am EDT Thursday Tropical Weather Outlook, NHC gave the disturbance 2-day and 5-day odds of development of 70%, respectively.

Bob Henson

Severe Weather Tornado

Atlantic's First Invest of the 2015, 90L, Organizing Over the Bahamas

By: Jeff Masters and Bob Henson , 3:50 PM GMT on May 06, 2015

The first Atlantic Ocean "Invest" of 2015 has arrived, as the National Hurricane Center (NHC) designated the area of disturbed weather over the Northwest Bahamas as Invest 90L on Wednesday morning. Note that there is no formal definition of what qualifies as an "Invest"; declaring an "Invest" is merely done so that a set of forecasting aids like computer model track forecasts can be generated for the disturbance. NHC gives an "Invest" a tracking number 90-99, followed by a single letter corresponding to the ocean basin--"L" for the Atlantic, or "E" for the Eastern Pacific. Other warning agencies assign "Invests" for the other ocean basins--"W" for the Western Pacific, "A" for the Arabian Sea, etc. When the numbering reaches 99, the next disturbance gets the recycled name "90". The appearance of 90L on May 6 this year marks the third earliest arrival of the year's first "Invest" over the past ten years:

2015: May 6
2014: June 4
2013: May 18
2012: February 5
2011: March 10
2010: May 24
2009: May 18
2008: May 31
2007: May 8
2006: June 10


Figure 1. Latest satellite image of Invest 90L.

Satellite loops show heavy thunderstorms between the Southeast coast of Florida and the Northwest Bahamas in association with 90L increased on Wednesday morning, but there was no evidence of an organized surface circulation trying to form. Long-range radar out of Melbourne, Florida showed no low-level spiral bands trying to form, and the activity was not well-organized. Wind shear was a moderate to high 15 - 25 knots. Water vapor satellite loops show a large area of dry air to the west of 90L over Florida and the Gulf of Mexico, and this dry air is retarding development, thanks to strong upper-level winds out of the west driving the dry air into the core of 90L. Ocean temperatures were near 26°C (79°F), which is about 1.7°C (3°F) above average for this time of year, and just at the limit of where a tropical storm can form. The Hurricane Hunter mission scheduled for Wednesday afternoon was cancelled, and has be re-scheduled for Thursday afternoon, if necessary.


Figure 2. Wind forecast for Saturday, May 9, 2015 at 11 am EDT made by the 00Z Wednesday run of the European model. The model is predicting a subtropical depression to be off the coast of the Southeast U.S.

Forecast for 90L
The 8 am EDT run of the SHIPS model predicted that wind shear over 90L would fall to the moderate range, 10 - 20 knots, on Thursday and Friday, which should allow 90L to approach subtropical depression status by Friday at the latest. Phase space diagrams from Florida State University from Wednesday morning's 06Z run of the GFS model support the idea that this system could be a subtropical or tropical system by Friday. Ocean temperatures fall to about 25°C (77°F) in the waters off of the North Carolina coast, so the farther north the storm wanders, the tougher time it will have developing tropical characteristics--though if the storm manages to find a sweet spot over the core of the warm Gulf Stream current, it has better odds of development. Steering currents are weak over the waters off the Southeast U.S. coast, so expect a slow and erratic motion for 90L. The Wednesday morning 00Z runs of our two top models for predicting tropical cyclones tracks, the European and GFS models, both showed the system making landfall this weekend, with the GFS model predicting landfall in South Carolina on Saturday, and the European model taking the storm ashore in North Carolina on Sunday. Beginning on Friday, coastal regions of both of these states can expect heavy rains and high surf causing rip currents and coastal erosion. Note that the west side of 90L will be weaker and drier, due to the dry air to the west of the storm, and the heaviest rains and stongest winds of 90L will be on the east side of the storm, over North Carolina. In their 8 am EDT Tropical Weather Outlook, NHC gave the disturbance 2-day and 5-day odds of development of 60%, respectively.

Wunderblogger Steve Gregory has a more detailed look at the meteorology of 90L in his Wednesday afternoon post.


Figure 3. Latest projected track of Typhoon Noul from the Joint Typhoon Warning Center (JTWC); a time of 9:00 p.m. JST Wednesday is 1200 GMT or 8 am EDT.

Typhoon Noul posing an increased threat to northern Philippines
After dumping more than 10 inches of rain at Yap International Airport, Category 1 Typhoon Noul is on its way toward the northern Philippines. Noel is gradually intensifying, with sustained winds of 90 mph reported at 8 am EDT (1200 GMT) Wednesday. Satellite loops show that the storm’s center is obscured by central dense overcast, but microwave data obtained by satellite shows that an open eye is already present. Noel is well structured and passing over warm waters, with only weak to moderate wind shear, so continued strengthening into a Category 4 storm is expected. Noel will encounter a trough as it approaches the Philippines, which may force the storm to recurve before making landfall there. However, model guidance has been trending westward, albeit with some uncertainty, and the official Joint Typhoon Warning Center track (see Figure 3) now brings Noul ashore along the northeast coast of Luzon, the largest and northernmost island of the Philippines. Noel is expected to peak in intensity just a few hours before approaching Luzon, with sustained winds of 105 knots (120 mph) projected by the JTWC near landfall. The northeast part of Luzon is mountainous, which could increase the risk for very heavy rains as Noul moves onshore or nearby, although a grazing landfall would put most of Luzon on the weaker western side of the circulation. Noul will be referred to as Dodong in the Philippines’ naming system.

Another tropical system, Invest 93W, is organizing to the east, and is likely to develop late this wee. It is too soon to know what its chances are of affecting the Philippines or Japan next week.

An exceptionally busy early portion of typhoon season
Noul's formation date of May 3 marks the second earliest appearance on record for the Northwest Pacific's sixth named storm of the year, according to statistics of the Japan Meteorological Agency's database from 1951 - 2015 maintained by Digital Typhoon. The average is 1.8 storms before May 8. The record is held by 1971, when the sixth named storm of the year (Babe) formed on May 3, six hours earlier than Noul's formation time. Noul will be the second tropical cyclone to affect the Philippines so far in 2015. The first was Tropical Storm Maysak, which hit the Philippines exceptionally early in the season--during Easter weekend, April 4 - 5. Fortunately, Maysak was weakening rapidly as it made landfall, and no deaths or significant damage were reported (though four people were injured after huge waves generated by Maysak hit them while they were taking selfies along the shoreline of Dipaculao town in Aurora province on April 4.)

Jeff Masters and Bob Henson

Hurricane

A New Tool for Gauging the Odds that a Storm Will Turn Severe

By: Bob Henson , 10:30 PM GMT on May 05, 2015

Will that thunderstorm become severe? It’s a question asked countless times each spring across the nation’s severe-weather corridors, not only by the public but also by forecasters in the National Weather Service. NOAA’s Storm Prediction Center issues convective outlooks that paint the broad picture of where severe weather is expected each day over the next week. SPC also coordinates with local offices on the placement of severe weather watches. It’s then up to local NWS forecasters to keep track of fast-breaking storm evolution and issue the county-scale warnings that millions of people act on. According to NWS severe storm services leader John Ferree, NWS offices issued more than 22,000 tornado warnings and more than 145,000 severe thunderstorm warnings in the seven-year period from 2008 through 2014.

Here at NOAA’s Hazardous Weather Testbed (HWT) in Norman, OK, I’m observing the first week of the annual Spring Experiment. The goal is to test promising new technologies that can advance the lead time and accuracy of warnings and “nowcasts.” Researchers from around the country will be filtering in and out of the HWT throughout the Spring Experiment. Each week, several forecasters from local NWS offices, plus one broadcast meteorologist, will also set up shop in Norman. These forecasters are issuing warnings with prototype tools just as if they were on an operational shift, so the pace is fast and intense, but these warnings don’t go out to the public, and the forecasters aren’t being graded on accuracy. Instead, they’re providing a running account of their experience—what worked best, what could be tweaked, and the like—in the form of live blogs with narratives and screenshots. According to experiment coordinator Gabriel Garfield (NWS/Norman), “We want the forecasters to be able to slip out of their comfort zones and think about how they can use probabilities in decision making.”

At the end of each week, the forecasters will summarize their work in a “Tales from the Testbed” webinar for colleagues in Norman and beyond. All this input will help refine the tools being studied, some of which will roll into regular NWS operations over the next few years.


Figure 1. Holly Obermeier (left), a broadcast meteorologist with Omaha’s KETV, confers with Steven Fleegel, a forecaster at the NWS office in Marquette, MI. Image credit: Bob Henson.


ProbSevere: A new window on storm evolution
The Spring Experiment is a sprawling, multifaceted project, as evident from the capsule descriptions on its webpage. Here’s one snapshot of a single tool now in its second year of testing: the ProbSevere model. It gauges the likelihood that a particular storm’s first instance of severe weather will occur during the next 60 minutes. Once a storm is established as severe, public awareness is easier to sustain, but anticipating that first appearance of high wind and/or large hail—before people on the ground know that a storm is severe—is crucial. The goal of ProbSevere is to give forecasters the equivalent of several radar scans of advance notice (about 10 extra minutes, on average) that a storm is approaching severe levels.

ProbSevere is based on a statistical blend of five measures linked to storm intensification:
—Instability (MUCAPE), provided by hourly updates from the Rapid Refresh forecast model
—Vertical wind shear, also provided by Rapid Refresh
—How quickly the top of the storm is rising, as deduced by satellite data
—How quickly ice crystals are forming (glaciation), also derived from satellite
—The diameter of the largest hailstones, as estimated by the Multi-Radar/Multi-Sensor System (MRMS).

Sometimes the environmental conditions will point toward severe weather before the MRMS values indicate severe hail, or vice versa. Since each storm follows a unique path to severity, ProbSevere relies on an equally weighted blend of the five indices above, rather than a simple “yes/no” verdict from any single index. The result is a single percentage-based probability that an initial severe report will occur in the next hour (see Figure 2). These probabilities can rise dramatically over just a few minutes, so forecasters at the experiment are watching for such trends in ProbSevere to gain more confidence that a severe thunderstorm warning may be justified. Other tools being tested this spring can also help make the call, such as an algorithm that detects a sudden jump in the total amount of lightning being generated by a storm. I’ll cover some more of these tools later in the week. Many are designed to take advantage of data from sensors aboard GOES-R, the first of the next-generation GOES satellites that will be deployed over the next few years. GOES-R is scheduled for launch in spring 2016.

ProbSevere was developed by NOAA and the Cooperative Institute for Meteorological Satellite Studies. You can check out the model for yourself at the ProbSevere website. The CIMSS Satellite Blog also has a helpful set of posts from the last year demonstrating ProbSevere through several case studies. A frequently updated subset of the daily liveblogs produced by forecasters at the Spring Experiment can be found at the GOES-R Proving Ground (HWT) blog.


Figure 2. At 4:08 pm CDT Monday, the NOAA/CIMSS ProbSevere model assigned an 89% chance that the storm shown here (outlined in magenta) in far northeast Kansas will produce its first instance of severe weather over the next hour. The probability is based on a statistical blend of the five indices shown in the text block next to the storm. These refer to instability, wind shear, hail size, the storm’s vertical growth rate, and glaciation (the pace of ice crystal formation within the thunderstorm). One of the forecast teams issued a non-public experimental severe thunderstorm warning at 4:17 pm CDT. No severe reports had come in by Tuesday morning for the time period of the warning, but in many cases severe weather in sparsely populated areas cannot be confirmed—and of course, even an 89% chance of severe weather means that no severe reports would be expected 11% of the time. Image credit: Kristin Calhoun, National Severe Storms Laboratory.

A week of deluge and severe weather
After years of intense drought, the plains of West Texas are soaking up one of their best episodes of spring rainfall in years. Several rounds of storms overnight produced 24-hour rainfall totals of 3 - 5” from the Lubbock area south toward Midland, and one cooperative observer in the town of Tahoka reported 9.10”. Four of the last six years gave Lubbock less than 13” of rain each, so apart from localized flash flooding, this is a much-appreciated dousing. As energy dives in and out of the upper-level low over the Southwest, the corridor of repeated storm development will shift only very slowly eastward over the next few days. This will keep the tap running from Texas across Oklahoma and Kansas as well as northeast Colorado, where atmospheric moisture is at near-record levels for this time of year.


Figure 3. NOAA’s Weather Prediction Center is calling for general rainfall totals on the order of 2 - 5” across much of north Texas, Oklahoma, Kansas, and northeast Colorado from Tuesday through Sunday. Image credit: NOAA/WPC.

Severe weather is a possibility each day this week, although the extensive rain and clouds will tend to tamp down the available instability. If early-day storms clear out in time, conditions appear ripe for more significant severe weather along a dry line in western Oklahoma and Kansas late Wednesday, perhaps extending into the central Plains on Thursday. A more widespread bout of severe weather is quite possible from Texas to Kansas by Saturday, as the main upper low begins to lift out. The likelihood of tornadoes should rise by that point. However, given that it’s May on the Great Plains—the equivalent of September in hurricane country—a stray tornado is possible on just about any afternoon this week. On Monday night, storm chasers from Australia caught images of a dramatic but nondestructive tornado near Garden City, KS (see Figure 4), and a Tuesday afternoon twister struck several miles southwest of Roswell, NM. However, the most dramatic tornado damage of the last 24 hours occurred in northern Germany, where the town of Bützow experienced significant impacts and a number of injuries.

Watching for subtropical development off the Florida coast
Models continue to indicate that a cluster of heavy showers and thunderstorms north of the Bahamas will consolidate into low pressure that may become a subtropical cyclone later this week. See the Jeff Masters post from earlier today for more details. Jeff will have a full update on Wednesday.

Bob Henson


Figure 4. This picturesque tornado was photographed by Clyve Herbert and Jane ONeill (Australian Sky and Weather). A tornadic signature was evident in Doppler wind data (upper left), and a hook appeared in the precipitation data (upper right). Graphic credit: NWS/Dodge City.

Severe Weather

Low Pressure Between Southeast Florida and Bahamas has Potential to Develop

By: Jeff Masters , 3:57 PM GMT on May 05, 2015

Heavy thunderstorms and strong winds are increasing over the waters between Southeast Florida and the Bahama Islands, where a non-tropical region of surface low pressure is developing. Radar out of Miami shows no organization to the shower activity, with rainfall amounts mostly less than an inch over Southeast Florida as of Tuesday morning. The first week of May is usually too early for the Atlantic to see its first named storm, but this area of low pressure could become Subtropical Depression One late this week, according to two of our top computer models for forecasting genesis of tropical cyclones, the GFS and European models. These models predict that the extratropical storm will form a well-defined center on Wednesday, then drift slowly northwards towards North Carolina during the week, gradually acquiring tropical characteristics. Ocean temperatures are near 26°C (79°F), which is about 1.7°C (3°F) above average for this time of year, and just at the limit of where a tropical storm can form. If the storm manages to find a sweet spot over the core of the warm Gulf Stream current, it has the potential to develop into a subtropical or tropical depression late in the week, as indicated by Phase space diagrams from Florida State University from Tuesday morning's 06Z run of the GFS model. Steering currents are weak over the waters off the Southeast U.S. coast, and the models are split on whether or not the disturbance will eventually make landfall on the Southeast U.S. coast late this week. I give the system a 40% chance of bringing heavy rains to the coast and high surf causing rip currents and coastal erosion late this week. In their 10 am EDT Tuesday Tropical Weather Outlook, NHC gave the disturbance 2-day and 5-day odds of development of 20% and 40%, respectively. The Hurricane Hunters are on call to investigate the disturbance on Wednesday afternoon, if necessary.


Figure 1. Wind forecast for Friday, May 8, 2015 at 2 pm EDT made by the 00Z Tuesday run of the European model. The model is predicting a possible subtropical depression to be off the coast of the Southeast U.S.

Bob Henson will have a post late this afternoon from the NOAA Hazardous Weather Testbed.

Jeff Masters

Hurricane

Subtropical Storm May Still Develop Off Southeast Coast This Week

By: Bob Henson , 3:23 PM GMT on May 04, 2015

Models are still suggesting that a moisture-laden disturbance off the southeast U.S. coast could develop into a subtropical cyclone later this week, although there is large uncertainty about how strong it might be and where it might go. A band of rich moisture now extends from the northwest Caribbean across Cuba into the southeast Bahama Islands. This air mass will shift north over the next several days, where it will intercept a decaying frontal boundary. Heavy rains will develop over the Bahamas over the next couple of days, possibly extending to the southeast Florida coast. If a closed low does form, it will likely be blocked from moving north as upper-level high pressure builds over the Northeast. The GFS model has been bullish on the prospects of a closed surface low forming along the boundary by midweek, then moving very slowly into the weekend. The 00Z and 06Z Monday GFS runs develop the this and bring it into the Georgia/South Carolina coast on Friday, with winds near the threshold for a tropical/subtropical storm, then keep it in that vicinity through the weekend. Such a scenario could bring several days of high surf and potentially heavy rain to parts of the southeast U.S. coast. The European model had been more tepid than the GFS, calling for a much weaker system located further offshore; the 00Z Monday run of the Euro is closer to the GFS in strength, but still keeps the system well offshore, moving away from the coast by the weekend. The 12Z Monday NAM model is even stronger than the GFS with a closed low projected to move close to the South Carolina coast by Thursday evening; however, the NAM is generally less reliable than the GFS or Euro in predicting tropical storm evolution.


Figure 1. The 90-hour surface forecast from the 06Z Monday GFS model run, valid at 8:00 pm EDT Thursday night, May 7. Image credit: WunderMap.


The low should initially develop as a result of the juicy low-level air mass being overtopped by a strong impulse in the powerful subtropical jet stream fed by the current El Niño conditions. Models show the high-altitude flow remaining cyclonic over the region, rather than the anticyclonic flow that would be associated with a typical tropical cyclone. However, winds at the center of the upper low will be quite light over the region of interest, which could allow for a compact warm-core circulation to develop. Sea-surface temperatures are also unusually warm--more than 2°C (3.6°F) above average over a large area just north of the Bahamas and east of Florida, close to where the system would be taking shape. Even so, these SSTs are still marginal for tropical-storm development, running at or above the commonly used 26°C threshold in a limited area from the Bahamas through a narrow band of Gulf Stream waters extending north toward the latitude of the Carolinas. The most recent phase-space diagrams from Florida State University, based on the 06Z Monday run of the GFS, show the system evolving as a very weak tropical cyclone (symmetric warm-core system), bringing it onshore late this week over South Carolina. In a special tropical statement issued at 10 am EDT Monday, the National Hurricane Center pegged the odds of at least subtropical development at 30% over the next five days, but near 0% through Wednesday. An Air Force hurricane-hunter aircraft is on call to investigate the system on Tuesday morning if necessary. If the low becomes a subtropical or tropical storm, it will be named Ana (pronounced "AH-na"]. Regardless of how the disturbance evolves, we can expect an increasing risk of high surf and rip currents from Florida to the Carolinas over the next several days.

Update: Wunderground blogger Steve Gregory weighs in on the potential subtropical/tropical system in his latest post. Readers may also enjoy the video update provided by Levi Cowan this afternoon at his Tropical Tidbits website.

On the scene at the NOAA Hazardous Weather Testbed
I’ll be blogging this week from the NOAA Hazardous Weather Testbed, located at the National Weather Center in Norman, Oklahoma. Each year the HWT and its Experimental Warning Program carry out a multi-week Spring Experiment that brings forecasters and researchers together to test-drive the latest in severe-weather warning technology. This year’s experiment begins today and runs through June 5. There should be at least a seasonable amount of severe weather across the southern Plains through this week, as ample moisture interacts with energy from an upper low gradually taking shape over the Southwest. The most active days appear to be Wednesday and Saturday right now. With widespread nonsevere showers and thunderstorms also expected, the trick will be anticipating where pockets of instability might develop after each night’s round of storms, and where small-scale ripples in the jet stream could set off a new batch of severe weather. For a project designed to push the boundaries of predictability, these challenges could be just the ticket.

Bob Henson

What Do Skyscrapers, Thundersnow, and Jim Cantore Have in Common?

By: Bob Henson , 6:40 PM GMT on May 01, 2015

Thundersnow is a rare enough event to get even veteran meteorologists like The Weather Channel’s Jim Cantore excited, as one can tell by the popular clips of Cantore reacting to several thundersnow events this past winter in Massachusetts. It’s estimated that less than 1% of all U.S. lightning flashes in the winter are associated with snow. Thundersnow’s rarity, and the fact that it’s often accompanied by poor visibility, has made it tough for researchers to learn much about it. However, some important leaps of progress have occurred in the last few years. Perhaps the most surprising insight: many cloud-to-ground lightning strikes observed in thundersnow are actually “ground-to-cloud” strikes, initiated by skyscrapers, wind turbines, and other tall objects.


Figure 1. Jim Cantore reports from Plymouth, Massachusetts, early on February 15, 2015, during a lightning-studded snowstorm. Image credit: The Weather Channel.


A study published last year in the Journal of Geophysical Research: Atmospheres examined the blizzard of February 1-2, 2011, which dropped more than a foot of windblown snow from Oklahoma to Michigan and trapped hundreds of vehicles along Chicago’s Lake Shore Drive. From his office in Rapid City, South Dakota, lightning researcher Tom Warner (ZTResearch) kept tabs on weather radar and lightning data during this storm, while also keeping an eye on The Weather Channel. As Warner recalled in an email: “I started noticing isolated lightning events in the snow sector, and when I plotted the coordinates of these events in Google Earth, they were consistently falling on objects such as tall radio/TV towers, tall buildings, and wind turbines. When the activity approached Chicago, I watched Jim Cantore live as he reacted to the lightning events. They were again grouped around the tallest buildings, which he was close to.”

After the storm, Warner scrutinized the data more closely with colleagues Timothy Lang (NASA Marshall Space Flight Center) and Walt Lyons (FMA Research/WeatherVideoHD.TV). Within the zone of heavy snowfall, they found a total of 282 cloud-to-ground and intracloud lightning flashes reported by the National Lightning Detection Network. The NLDN’s array of energy-sensing instruments can track the evolution of a lightning flash from millisecond to millisecond and pin down the flash’s location to within 500 meters. The researchers then mapped the NLDN data against the locations of communication towers, tall buildings, and other towering structures. Of the 282 flashes, they found that 72% occurred within 1 km of a tall object, and most of the other flashes were within 3 km of such an object. Many of the flashes reported in the Chicago area were distinctly clustered around particular structures, such as the Willis Tower and Trump Tower. At the same time, there was very little lightning detected over the nearby waters of Lake Michigan, although radar showed heavy snow falling there.


Figure 2. Lightning flashes detected by the National Lightning Detection Network in downtown Chicago during the February 1-2, 2011, blizzard. Tight clusters of negative flashes (-CGs and -ICs) are evident around the Willis Tower (1729 feet to its tip) and Trump International Hotel and Tower (1389 feet). These flashes are related to lightning channels that initially propagate up from the towers. No lightning was reported from the nearby John Hancock Center, even though it is in the same height range (1506 feet). Image credit: Journal of Geophysical Research/American Geophysical Union.


The two ways lightning can travel upward
As its name implies, cloud-to-ground lightning strikes (CGs) involve electrical charge being lowered from the base of a thunderstorm to an object at ground level along a zigzag channel called a leader. A much brighter flash, the return stroke, occurs after the leader connects to the ground, when charge accelerates downward through the leader but in an upward-cascading fashion (somewhat like a cleared traffic jam suddenly freeing cars further back along the highway). This process can repeat itself more than a dozen times in less than a second, with lightning photos often revealing a forked structure due to new leaders that branch outward and downward from the main lightning channel.

In a typical summer storm, cloud-to-ground lightning will often strike the tallest object around--an isolated tree or house, perhaps--but the data suggest something else was going on in Chicago. When an tower is at least several hundred feet tall, the odds increase that the initial leader will actually propagate from the top of the tower up to the cloud base, rather than the other way around. Just as a downward-propagating flash can branch out over its brief lifetime, striking several points on the ground, an upward-propagating flash can branch out as well (see Figure 3 below). This may “trick” the NLDN into reporting several closely clustered ICs or CGs (as in Figure 2) instead of a single, upward-pointing flash.

Upward lightning has been recognized since the 1930s, but high-speed cameras, sophisticated electric field meters, and 3-D lightning mapping tools have greatly advanced our ability to study it. Over the last few years, scientists have analyzed upward-directed lightning at several locations, including a set of 10 radio/TV towers near Rapid City. A total of 67 upward flashes originated from these towers from 2012 through 2014, when a field project called UPLIGHTS was carried out. More than a third of those flashes (25) occurred during the intense and destructive winter storm of October 2, 2013. Prior to UPLIGHTS, 81 upward flashes were observed in the Rapid City area from 2004 to 2010. Almost every upward flash in this period occurred less than half a second after a strong CG lowered positive charge to ground less than 30 miles away. It appears the positive CGs intensify the local electric field enough to generate a compensating upward flash from a nearby tower, where charge can be readily concentrated. This is referred to as lightning-triggered upward lightning (LTUL).


Figure 3. Upward lightning from four towers in Rapid City, SD. Tall structures appear to be the source of many of the lightning flashes observed during snowstorms. (Photo © Tom A. Warner, used with permission.)


Research in South Dakota and Alabama, as well as Japan and Austria, has identified a separate phenomenon called self-initiated upward lightning (SIUL), which almost always occurs during snowstorms with low cloud bases and low freezing levels. Warner, Lang, and Lyons weren’t able to deploy the same observing tools for the February 2011 storm that were used in UPLIGHTS, but a variety of evidence strongly implies that the lightning in this snowstorm was mostly in the form of SIULs. Interestingly, SIULs seem to require winds of at least 18 mph at the top of the tall object--except when that object is a spinning wind turbine. It appears that either the wind or the motion of the turbine can dispel a protective region of coronal discharge that would otherwise inhibit the upward development of a lightning flash. With that protection gone, the enhanced electric fields at the top of the tall structures appear to be enough to generate a leader that can make it up to the regions of opposite charge in the relatively low nimbostratus clouds.

Why don’t we get more “regular” lightning in thundersnow?
One possible reason that typical cloud-to-ground strikes aren’t very common in thundersnow is the relative weakness of the convection (the upward motion within the snow-producing cloud). Cloud-to-ground lightning originates from the intense pockets of charge that develop in a thunderstorm as snowflakes, ice pellets (graupel), and unfrozen water droplets jostle each other. Radar analysis for the February 1-2 storm showed that almost every flash of thundersnow occurred where the peak radar reflectivity was less than 30 dBZ, a value normally considered too low to produce lightning. However, in most cases there were higher reflectivities up to 30 miles upstream. This suggests that charge may have been generated within small convective cells, then carried downwind by droplets and crystals. Chris Schultz (NASA) and colleagues have carried out detailed analyses of several upward lightning flashes in the vicinity of Huntsville, Alabama. One such flash, which occurred on January 10, 2011, appears to have propagated from a TV transmission tower into a cloud that featured sloped layers of mixed ice crystals, ice pellets, and supercooled water. The researchers have similar data for flashes from tall towers near Washington, D.C., and Baltimore, Maryland. “Hopefully we can get back to these cases and understand the radar measurements better, to infer what is going on microphysically within the cloud at the time of the flash,” says Schultz.

In summary, tall towers don’t seem to be an absolute requirement for thundersnow, but they do appear to facilitate it in some ways that are only now becoming more apparent. The JGR study found that even objects less than 300 feet tall, such as cellphone towers and transmission lines in rural areas, can be enough to trigger thundersnow. Patrick Market (University of Missouri) is a longtime expert on the climatology of thundersnow. In an email, he told me: “The literature supports the occurrence of thundersnow before the advent of radio, so we know that the Earth's atmosphere is capable of the phenomenon all on its own. That said, the evidence that has been presented for taller structures influencing CG activity is compelling. The outstanding question that is difficult to quantify is: Are there more flashes in winter thunderstorms because of taller man-made structures?”

What's Up With This? Yes, this lightning does appear to be striking upward from a tall TV tower. Turns out that a small number of lightning strikes can be called "upward lightning" as the initial leaders are from the ground-to cloud. And these events appear also to occur during blizzards and may account for many reports of "thundersnow." (c) Tom Warner/WeatherVideoHD.TV. For more info see: https://www.weathervideohd.tv/wvhd.php?mod=search&sev=36677&sp=1

Posted by Weathervideohd.tv on Wednesday, April 29, 2015

Video 1. A classic “ground-to-cloud” lightning flash captured by high-speed photography (1000 frames per second) at Rapid City, South Dakota, on June 23, 2010. Bright, short-lived return strokes can be seen near the top at some distance from the main lightning channel. If embedded video does not load, you can find it at the WeatherVideoHT.TV Facebook page (posting date is April 29, 2015) or on their previewing site. Video credit: Tom A. Warner/WeatherVideoHD.TV.


Atlantic’s first named storm still possible next week
This morning’s round of model output continues to indicate the possibility of a named subtropical or tropical cyclone near the southeast U.S. coast toward the latter part of next week. See Jeff Masters’ post from earlier today for details. We’ll keep an eye on this potential system and will have an update by Monday at the latest.


This week’s WunderPoster: Fallstreak hole
The first phase of our WunderPoster series concludes today with a strange feature called a fallstreak hole, also known as a hole-punch cloud. These most often form when an airplane ascends or descends through a shallow layer of cloud made up of supercooled water droplets (droplets that remain liquid at temperatures well below freezing, due to a lack of particles on which to form ice). The local pressure drop just behind the aircraft’s propellors or wings can generate enough cooling to produce a batch of fast-growing ice crystals that consume moisture from the surrounding droplets, leaving a mid-cloud hole.

All of the 13 WunderPosters to date can be downloaded in formats suitable for posters or postcards. Watch this space in May for the release of a new set of WunderPosters inspired by photos contributed by the WU community.

Bob Henson

Lightning Winter Weather

An Atlantic Named Storm Coming for the First Week of May?

By: Jeff Masters , 2:07 PM GMT on May 01, 2015

The first week of May is usually too early for the Atlantic to see its first named storm, but that is a possibility this year, according to the Friday morning runs of the GFS and European models. These models predict that an extratropical storm will form along an old cold frontal boundary over the Bahama Islands just east of the coast of Florida on Tuesday, then drift slowly northwards towards North Carolina during the week. Ocean temperatures are near 26°C (79°F), which is about 1.7°C (3°F) above average for this time of year, and just at the limit of where a tropical storm can form. If the storm manages to find a sweet spot over the core of the warm Gulf Stream current, it has the potential to develop into a subtropical or tropical depression in the May 7 - 8 time frame. Phase space diagrams from Florida State University from Friday morning's 06Z run of the GFS model support the idea that this system could be a subtropical or tropical system until it pushes north of a line even with the South Carolina/North Carolina border, where ocean temperatures fall to about 25°C (77°F). It is too early to put odds on whether or not this storm might make landfall, but South Carolina and North Carolina might see some rain from this system by Thursday.


Figure 1. WInd forecast for Thursday, May 7, 2015 at 8 am EDT made by the 00Z Friday run of the European model. The model is predicting a possible subtropical depression to be off the coast of the Southeast U.S.

Later today, Bob Henson will have a fascinating post on how human structures may be causing winter lightning.

Jeff Masters

Hurricane


The views of the author are his/her own and do not necessarily represent the position of The Weather Company or its parent, IBM.

Category 6™

About

Cat 6 lead authors: WU cofounder Dr. Jeff Masters (right), who flew w/NOAA Hurricane Hunters 1986-1990, & WU meteorologist Bob Henson, @bhensonweather