Published: March 24,2015
Number
of days each year with at least one (E)F1 tornado (black squares) and
more than 30 (E)F1 tornadoes (gray triangles) from 1954-2013. Average
over each decade indicated by large dots and line plots.Fewer Tornado Days, But More Active Days
When eliminating (E)F0 tornadoes from yearly counts, which have steadily risen over the past few decades due to more extensive spotter networks, the implementation of Doppler radar, and advanced technology such as smartphones and social media, there is essentially no long-term yearly trend in the raw number of (E)F1 and stronger tornadoes.(MORE: The Enhanced-Fujita Scale)
However, the number of days with at least one (E)F1+ tornado in the U.S. has fallen from an average of 150 such days in the early 1970s to around 100 days in the first decade of the 21st century, according to an October 2014 study in the journal Science.
However, the study by noted tornado researchers Dr. Harold Brooks of the National Severe Storms Laboratory, Greg Carbin of NOAA's Storm Prediction Center, and Dr. Patrick Marsh, also of NOAA/SPC, found the number of days with a large number of tornadoes is actually increasing over time.
"The frequency of days with more than 30 (E)F1+ (tornadoes) has increased from 0.5 to 1 days per year in the 1960s and 1970s to 3 days per year over the past decade," says the Brooks et al. study.
In essence, we have fewer days with tornadoes, but are packing more of them into the days we have. "Approximately 20 percent of the annual tornadoes in the most recent decade have occurred on the three biggest days of each year," says the Brooks et al. study.
Another recent study by Dr. James Elsner not only found a similar clustering of tornadoes into fewer days, but also a spatial clustering of tornadoes on those very active days.
"It appears that the risk of big tornado days with densely concentrated clusters of tornadoes is increasing," Elsner says in the July 2014 study.
(Silver Lining Tours)
Large Swings in Monthly, Yearly Numbers
For only the second time since 1950, the first three weeks of March 2015 passed without a single tornado anywhere in the U.S.(MORE: 2015 Severe Season Reaches 'Uncharted Territory')
Yet as recently as 2011, almost 1,700 tornadoes ripped across the nation, including 349 tornadoes in a four-day outbreak from April 25-28, the costliest tornado outbreak in U.S. history.
(MORE: 10 Shocking Things About the April 2011 Outbreak)
While year-to-year variability has long been prevalent in U.S. tornado counts, a 2014 study by Dr. Michael Tippett found volatility, a term he uses for variability in tornado counts, has increased since 2000.
Furthermore, the Brooks et al. study found the tendency for more monthly extreme highs and lows in (E)F1+ tornado counts in recent years.
"Excluding the zero-tornado months, there are more extreme months in the most recent 15 years of the database (1999-2013) than in the first 45 years," says Brooks et al. 2014.
In other words, we've seen extreme high monthly tornado counts (758 tornadoes in April 2011, for example) and extreme low monthly tornado counts (March 2015, for example) more often over the past 15 years, a trend that may continue.
Of course, low tornado count years do not preclude significant tornadoes or tornado outbreaks. Despite the lowest three-year tornado count on record from 2012-2014, we still had destructive outbreaks in March 2012, in May 2013 (Moore and El Reno, Oklahoma), and April 2014 (Vilonia, Arkansas).
Julian date of the 50th (E)F1+ U.S. tornado each year from 1954-2013.Skewing Earlier in the Year
Tornadoes can occur any time of year the overlap of sufficient moisture, atmospheric instability -- relatively cold, dry air aloft overlying warm, humid air near the Earth's surface -- and a strong source of lift such as a warm front, dryline, strong jet-stream disturbance occur.Because of that, it's difficult to define a tornado season on a national scale as distinctly as, say, a hurricane season.
However, Brooks et al. tracked as a metric the occurrence of the year's 50th EF1+ tornado to get a sense of whether the timing of the ramp-up in U.S. tornadoes typically seen in spring is changing.
While the long-term average date (March 22) hasn't changed, Brooks et al. found a marked increase in the number of "late-start" and "early-start" years since the late 1990s. The four latest starts and five of the ten earliest starts to the season all occurred in the 1999-2013 period. These range from late January (1999 and 2008) to late April (2002, 2003, 2004 and 2010).
In essence, even the date the season kicks into a higher gear is becoming more volatile.
Climate Change Role?
Now, the toughest question: Is climate change playing a role in the increasing variability of the nation's tornadoes?The short answer is, possibly.
The challenge in answering this question is linking short-fuse events like tornadoes and tornado outbreaks to long-term changes in atmospheric parameters generally conducive for severe thunderstorms, such as instability and wind shear.
Studies by Dr. Jeff Trapp and Dr. Noah Diffenbaugh, among others, suggest atmospheric instability, driven by increased moisture, is expected to be greater in a warming climate. However, wind shear, crucial for the formation of supercells which can produce the strongest tornadoes, may diminish overall, but may feature more days with higher wind shear.
Therefore, the overall environment may be more conducive for severe thunderstorms (with large hail and damaging winds), but it remains unclear whether the number of tornadoes or even strong tornadoes would necessarily rise in a warming world.
This brings up an interesting possibility, a seasonal outlook for severe weather, similar to hurricane season outlooks.
(MORE: El Nino - Tornado Season Link?)
"I suspect that ultimately knowing if a severe weather season will be above, below, or near normal would be important for reinsurance portfolios as an increasing amount of money is spent on hail and wind claims," said Dr. Patrick Marsh from NOAA/SPC.
No comments:
Post a Comment