Severe Thunderstorms, Tornadoes, Lightning All Down in 2010

UPDATE 4/21/10: We have written a companion news story to this blog entry, which goes deeper into the discussion about the cool Gulf of Mexico. Below is the latest map of sea-surface temperature anomalies there.

Another way to measure the severity of the weather is storm Risk maps from NOAA's Storm Prediction Center. We are already one month past their record late "Moderate Risk."

ORIGINAL REPORT:

No matter how you slice it, the 2010 United States severe weather season is much slower than last year, which was much slower than 2008. This is true whether you look at the number of lightning strikes, tornado (and all severe) weather reports, or even watches and warnings issued by NOAA.

Looking specifically at the Spotter stats, these numbers come from people calling the National Weather Service to report hail, wind or tornadoes. The map above compares January - March of 2008 and 2010; I have updated the numbers in the graphs below to cover January 1 - April 18 (the difference is mainly that the downward progression is more linear at the end of March because 2009 did a lot of catching up during the first half of April).

That's over 6,000 severe weather reports by this time of year in '08 but only 1,400 this year, and almost 500 tornadoes versus less than a hundred! But before moving on to more statistics, let us consider the pros and cons of this slow severe season. I believe there is a direct correlation between the amount of severe weather and lives lost; this is good news indeed for 2010 so far.

Though there is no easy way to measure it, I believe it's reasonable to project that there has also been less property damage (from hail, winds or lightning) -- good for both homeowners and the insurance industry, but not so good for roofers and other damage experts who rebuild or repair damaged homes and automobiles. Utilities, such as electric companies whose lines taken damage and energy companies, whose tanks (and gas lines) are set ablaze, are probably happy about the reduced lightning.

It's hard to relate cloud-to-ground lightning strike statistics (which is what I present below) to cloud-to-plane strikes but my feeling is that less planes have been struck by lightning this year, certainly it would seem there has been less news about it. If so, that's a good thing.

The only other people who are probably bummed about the lack of severe weather are severe weather enthusiasts and photographers (like myself), and storm chasers like the government's VORTEX2 project which is supposed to study tornadoes but didn't get much of a chance last year. Does a slow April mean a slow May for tornadoes? NOAA's Greg Carbin from VORTEX2 looked at the stats and didn't find a correlation, which gives them some hope, though AccuWeather.com believes that the tornado season may be displaced to their east (compare our tornado season forecast map versus their chase area map).

Without further ado, lets look at the stats. First, every time a lightning strike hits the ground in the U.S., Vaisala records it. Their stats for the last three years (through April 13th) are shown below:

Again the difference between 2008 (over 2.2 million strikes) versus this year (a mere 800,000) is clear to see. The reasons for both this year and 2009 being radically lower than 2008 are actually a little different. Last year cool air covered much of the East through June, preventing thunderstorms from developing. This year, the cool air cleared off the land in early March but the energy in the Gulf of Mexico has not been available to generate large severe weather outbreaks. This is because of the low dew points there (storms need rich, warm moisture to get started).

Michael Utley of Struckbylightning.Org says that there have been 8 people injured by lightning and none killed so far this year, versus 12 injuries and 1 death in 2009, though those stats are unconfirmed for this year. If you want to get theoretical, we can use the stats here to make the assumptions that (on average) there should have been 19 injured with 6 killed year-to-date for 2008 but only 7 injured and two killed this year so far. Either way, less lightning strikes is going to mean less casualties which is a good thing.

The National Interagency Fire Center statistics year-to-date also match the lightning data above, if you look at acres burned. However, only 12% of wildfires are caused by lightning and there is no correlation when you look at the number of actual wildfires, so this could be a coincidence.

Another stat we can look at is the number of weather watches (specifically, Tornado & Severe Thunderstorm) issued by NOAA's Storm Prediction Center. You'd expect these to follow a similar trend because a watch is only issued when widespread severe weather is expected.

But this is one interesting differentiator -- 2008 and 2009 were not that different, though there has been a big drop this year. This would seem to say that SPC was being more cautious last year than they are now - possibly because they kept underestimating the power of the cold air for holding back severe weather last year -- but I'm only conjecturing here.

Now onto warnings, which are issued by local National Weather Service offices. This is probably the least useful set of data because it adds in Flood and Marine warnings to Tornado and Severe Thunderstorm warnings, and because warnings are issued in different sizes for different parts of the country. Also, the criteria for warning for severe thunderstorms has been in flux during the past several years, with the NWS changing their minds on how large hail has to be to issue a warning.

Still, the trend is evident. Less than half the warnings have been issued this year vs 2008, and only 1/4 of the Tornado Warnings were issued, which is a more important number. We can also look at the area warned for Tornadoes, which makes the data more valuable (because warnings are different sizes in different weather situations and geographical areas).

This is one of the best comparisons that I can find that suits the public -- it speaks to how large of an area has been under a tornado threat this year and again, it's minuscule compared to previous years.

But how do we stack up further back in history? As you may recall, February almost tied a record for the least number of tornadoes for that month going back to 1950. Even through April 15th, we are very close to the minimum number of tornadoes year-to-date in recorded history -- at 79 versus 74 in the weakest year. (This number is "inflation adjusted" to account for changes in reporting methodology).

The raw number of tornadoes per year is also less than half of the weakest year since 2005. And as far as severe storm reports, if we take that graph above and back it up to 2000, you can see that the number of reports (like tornadoes) is much lower than anything we've seen recently.

Note that, unlike the graphs above, this shows that the severe weather trend has not been falling off a cliff like the 2008-2010 graphs might make it appear, in fact 2008 stands out as a banner year for severe weather reports. However, 2010 year-to-date is much lower than anything else on the graph. It wouldn't surprise me if it were the lowest number in the modern reporting era (since 1995) if not further back, but calculating that would involve a longer chunk of time than I've dedicated to all of the stats above.

*METHODOLOGIES: Source of lightning death/injury stats is unknown and it's impossible to say if "flash" is defined the same was as Vaisala does - hence use with extreme caution. Other notes: Graphs by Excel 2007. Data is "year-to-date" for March 31st, or April 18th as specified above, with the exception of Lightning (4/13) and Wildfires ("week of 4/23" which is presumably 4/18). Watch, spotter & tornado stats from SPC. Certainly not all events are called in as spotter reports, but major ones are and the last 10 years are comparable with each other (before the Internet era, the number of reports drops off rapidly due to a smaller number of spotters). Warning stats from IEM. I concentrated mainly on 2008-2010 because I had stats across the board for those years, and because I think that's about the limit of the public's short-term weather memory.

Report a Typo