I came across some recent research that the Storm Prediction Center placed on their website. Right away something jumped off the map to me. The Charlotte metro area appeared to have more severe weather days than any other location in the U.S. over the time period of 2003-2012.
2 things jump off these maps the overall severe weather reports and the damaging wind reports.
Here’s the description of the study and mapping methodology listed on the SPC website.
These maps were created by gridding the total number of daily severe weather reports (Midnight to 11:59pm CDT) over the 10-year period from 2003 through 2012 on a 80km grid. The resulting grid numbers are then divided by 10 and smoothed to arrive at the annual average number of days with a report of severe weather based on official NWS Storm Data records. The 80km grid-point value corresponds to the number of events within 25 miles of a point. While the resulting maps generally match our understanding of severe weather climatology, there are a few exceptions that come about as a result of how the severe weather event is quantified. Even through the data are smoothed, severe weather reports cluster around population centers. This can be seen on the “any” or “all” severe weather map in the upper left. Maximum values show up around Charlotte, NC, Huntsville, AL, Jackson, MS, Springfield, MO, and Dallas, TX. These are locations where more severe weather is reported because more people live in those areas. The hail reports used to generate the hail frequency map are from reports of hail 1 inch or greater in diameter. Large hail reports are most common from Rapid City, SD to Denver, CO, Dodge City, KS, and Springfield, MO. The wind report map perhaps poses the greatest challenge in terms of representing where a greater severe thunderstorm wind threat may exist. The majority of severe thunderstorm wind reports are verified by falling trees *not* by observed wind gusts of 50 knots or greater. Thus, there is a distinct tendency for severe thunderstorm winds to be reported in areas with more trees. Recent peer-reviewed studies have compared the severe weather reports used to make these maps with automated observations (for wind), and radar data (for hail). These studies have found that greater concentrations/frequencies of severe hail (based on radar) and 50kt or greater severe thunderstorm wind gusts (based on automated observations) are more likely to occur over parts of the Great Plains and Midwest than what might be indicated in the maps shown here. (Click each map for full resolution version.)
An Objective High-Resolution Hail Climatology of the Contiguous United States
John L. Cintineo, Travis M. Smith, Valliappa Lakshmanan, Harold E. Brooks, Kiel L. Ortega
Weather and Forecasting
Volume 27, Issue 5 (October 2012) pp. 1235-1248
Measured Severe Convective Wind Climatology and Associated Convective Modes of Thunderstorms in the Contiguous United States, 2003-09
Bryan T. Smith, Tomas E. Castellanos, Andrew C. Winters, Corey M. Mead, Andrew R. Dean, Richard L. Thompson
Weather and Forecasting (Also here.)
Volume 28, Issue 1 (February 2013) pp. 229-236
So what’s going on here?
While we do get many severe weather events or straight line wind damage. There are 2 things that are skewing the results over our area.
#1 Population: We have a very high population density. The more people you have the more severe weather reports you get. The population bias is an observation bias. The more eyeballs you have the more reports you get. Throw in smartphones and social media in our region and these reports get relayed quickly.
#2 Trees! The study states that most of the wind damage reports were of trees down not of actual wind speeds measured. As anyone from our area knows we have lots and lots of trees around here. So coupled with the population bias you can see how these numbers get inflated.
While we certainly get a lot of damaging wind events the magnitude of reports appears to be inflated by both population and tree bias. I love to see a map that correlated wind damage reports with Bradford Pear tree growth. I bet there would be a high correlation.