Severe Storm

Table of Contents

The National Weather Service defines a severe thunderstorm as a thunderstorm that produces a tornado, winds of at least 58 mph (50 knots), and/or hail at least ¾” in diameter. Structural wind damage may imply the occurrence of a severe thunderstorm. A thunderstorm wind equal to or greater than 40 mph (35 knots) and/or hail of at least ½” is defined as approaching severe. 

Background

What causes a thunderstorm?

Thunderstorms are typically caused by convection that occurs when the sun has heated a large body of moist air near the ground. This air rises and is cooled by expansion. The cooling condenses the water vapor present in the air, forming a cumulus cloud aggregation of minute particles of water or ice suspended in the air. If this process continues, the summit of the cloud often attains a height of 4 miles above the base, and the top spreads out in the shape of an anvil.

The turbulent air currents within the cloud cause a continual breaking up and reuniting of the raindrops, which may form hail, and builds up strong electrical charges that result in lightning. As the storm approaches an area, the gentle flow of warm air feeding the cloud gives way to a strong, chilly gust of wind from the opposite direction, blowing from the base of the cloud. Intense rain begins and then gradually diminishes as the storm passes.

Types of thunderstorms

  • Ordinary cell: As the name implies, there is usually only one cell with this type of thunderstorm. Also called a “pulse” thunderstorm, the ordinary cell consists of a onetime updraft and one time downdraft. 
  • Multi-cluster cell: While each individual thunderstorms cell, in a multi-cell cluster, behaves as a single cell, the prevailing atmospheric conditions are such that as the first cell matures, it is carried downstream by the upper level winds and new cell forms upwind of the previous cell to take its place.
  • Multi-cell line: Sometimes thunderstorms will form in a line which can extend laterally for hundreds of miles. These “squall lines” can persist for many hours and produce damaging winds and hail.
  • Supercell thunderstorms are a special kind of single cell thunderstorm that can persist for many hours. They are responsible for nearly all of the significant tornadoes produced in the U.S. and for most of the hailstones larger than golf ball size. Supercells are also known to produce extreme winds and flash flooding.
  • Straight line winds are winds that come out of a thunderstorm. To be classified as severe, they must exceed 58 miles per hour. 

Hazard clarification

For the purpose of KIPDA’s analysis, only wind is included within the severe storm analysis. Hail is analyzed as its own hazard, and heavy precipitation is analyzed under flooding. 

Cascading effects of severe storms

• Lightning

• Flash floods

• Hail

• Outflow

• Tornadoes

• Winds 

Location + Extent + Past Events

KIPDA used the NCEI database to find past high wind and thunderstorm wind events between 2000 and 2020[1]. Use the filters below to select event type – thunderstorm wind or high wind – and county. To view the data in spreadsheet form, click view data here.

*Magnitude is estimated or measured wind gust (MPH).

Probability

Occurrence

In the past 21 years, the KIPDA region has experienced 198 thunderstorm and high wind events according to the NCEI database. 

Please note that thunderstorm or high wind events that happened on the same date are classified as the same event. 

  • Bullitt County: 79 high or thunderstorm wind events/21 years = 3.8 wind events annually
  • Henry County: 58 high or thunderstorm wind events/21 years = 2.8 wind events annually
  • Oldham County: 77 high or thunderstorm wind events/21 years = 3.7 wind events annually
  • Shelby: 84 high or thunderstorm wind events/21 years = 4 wind events annually
  • Spencer: 45 high or thunderstorm wind events/21 years = 2.1 wind events annually
  • Trimble: 45 high or thunderstorm wind events/21 years = 2.1 wind events annually

Climate Change

Severe storms are a combination of many different factors, such as precipitation, wind speed, and hail, which are all individually affected by climate change. Due to the higher atmospheric temperature, the air can hold more moisture and precipitation is likely to be increased in some regions. The occurrence and speed of wind is likely to rise, as well, according to a 2013 study that predicts that non-tornadic wind events will increase in frequency due to climate change. [1] The effect on hail and hailstorms is more regional specific, but the general consensus is that the size of hail will increase. The amount of research done on the effect of climate change on severe storms is limited, however, so it is uncertain exactly how that will be affected. [2]

For the Midwest, scientists expect rising precipitation to be the most concerning factor of increased severe storms, as they are mostly expected to occur in the spring. Violent winds appear to have increased for a time and now are decreasing again, but they could possible speed up again in the future as a result of climate change. [3]

The Kentuckiana region is specifically expected to have more winter and spring precipitation, so severe storms are also predicted to increase during those seasons. The effect of climate change on summer and fall events is more uncertain. Evidence on how climate change will affect the type of severe storms that produce tornadoes is also limited. [4]

Overall Probability

Overall, the KIPDA HMP ranks the probability of a hail event for the region as extremely high because of its high annual frequency. 

Impact + Vulnerability

Property Data

The dashboard below shows residential property damage claims related to wind in the KIPDA region between 2010 and 2020.  These claims predominately cover single family and multi-family dwellings; however, a few commercial properties may be contained within the data. KIPDA obtained this anonymized data from Verisk’s Exactware program, which runs ExactAnalysis. ExactAnalysis is the largest full-cycle claims management software and reporting tool in the property insurance industry. In the past, hazard mitigation plans have used estimates from NCEI; these figures rely on local reporting and do not typically capture property loss. Therefore, this data from the ExactAnalysis program has provided a more robust data set on the true cost of severe storm events.

Because this data has never been used for hazard mitigation planning before, KIPDA’s GIS analyst conducted extensive cleaning. To learn more about this process, click the data cleaning toggle below. 

Claims Data Cleanup Actions Taken 

Sort Ascending by Value; delete all 0 and negative (-) values in “Gross Estimate Amount” column (10924 rows had Zero for the amount 22.5% of total data) 

Many suspect values remain ($1, etc.  basically anything less than a deductible (typically $500 or  

$1000) is suspicious to me and seems like that can’t be accurate (another 3320 have values under $500 and 6110 less than $1k) 

Look at numeric “Type of Loss”  able to ID that 104 – Wind and 150 – Water… need other numeric codes to ID type of hazard and make those rows useful… (Deleted all other numeric values (26 rows) for now, can restore from clean dataset later if desired) 

Assumed GL69 = General Liability  

Deleted “Type of Loss” 111-Theft(1); “All Other/ All Other Peril Not Enumerated/All other Perils/ All Other Physical Damage/ AOPD/APLWTR-APL (assumed this is appliance?)” (276) because doesn’t help our purposes if not associated with a NATURAL hazard; “Backup of sewer or drain (Sump Pump/Faulty Drain)”(10) 

Skipped to Cause of Loss column because those details may be more helpful at initial elimination of non-weather related claims: Deleted “pipe burst …..”(1); “accidental breakage by insured”(1); “all other(FIRE)” because no wildfires that year; “animals/ insects” (1); “appliance”(12); “auto driven by nonresident”(1); “Back-up sewer drain/ bathroom sink clogged/ bathtub related/ broken pipe”(131); “Theft/ Burglary” (11); “candle” (6); “car ran through and destroyed(impact by vehicle)”(1); “cigarette smoking” (2); “fire sub loss” no wildfires in 2017 (1); “water damage toilet malfun”(1); “COL: water Sub Loss: escape from ap” (sounds like appliance?)(1); “contamination/pollution” (1) doesn’t specify source so  unsure if natural hazard related; “cooking(FIRE)” (35); “Damage from vehicle – insured owned”(1); “Damage to Dwelling(Vandalism)”(5); “Damage to bathroom flooring(1)”; “damages to Claimant vehicle(CPL?)”(1); “Defective Appliances”(6); “Drain Bck (Drain backup??)”(4); “Dryer(FIRE)”(1); “Electric Baseboard Heater”(1); “Electrical Fire”(8); “Equipment(WATER)” (5); “Falling Objects”(1) unspecified source?; “Faulty Drain”(3); “Faulty Wiring”(20); “Fire”(20) verified date/location didn’t match wildfire data; “Fire-All other/Fireplace/Grease Fire”(3); “Fire Caused by Appliance”(5); “Fire Electrical” (1); “Fire External Source”(1)”; “Fire Total” (1);  “Fireplace/SolidFuel/Chimney” (21); “Fire Erupted From Brush Hazard” (4) verified does not coincide with wildfire dates; “FNOL….. (Water/ Wind)” (4); “Frozen Pipe (Weather Related – No/ Non-weather)” (4); “Fungi/Bacteria”(1); “Furnace”(3); “Grease Fire in Kitchen”(13); “HVAC” (1); “int and ext damage (VANDALISM)” (1); “Jewelry Missing from home” (2); “Kitchen/ other than grease (FIRE)” (7); “leaking pipe…”(1); “Light(OTHER)” (2); “Light(WATER DAMAGE – APPLIANCE)”(1); “Moderate(OTHER/UNKNOWN)(1); “mold general/in kitchen(MOLD/OTHER)” (2); “Nearby Fire (SMOKE)”(2); “None(FIRE)”(2); “off Premises/ on premises (THEFT)”(11); “Other(BACKUP OF SEWER OR DRAIN)”(1); “Other(BURG/BURGLARY THEFT)”(92); “Other(COLLAPSE)”(1); “Other (COLL-VEH)”(5); “Other(DAMAGE BY VEHICLE WATERCRAFT)”(1); “Other(DRAIN BCK)”(8); “Other(FALLING OBJECT)” (9) Verified not on same date as a quake; “other(FIRE)”(85) Verified not on/near date/location of wildfire, (1) entry excluded from deletion because close to date/location of a wildfire; “Other (FIRE-APPL/COOK/ELEC/FIRPL/GASLK/OTHER/PORT/SMOKE/VEH/UNDT”(43); “Other(HURRICANE)(1)”; “Other(LIABILITY)”(1); “Other(MISCHIEF)”(1); “Other(MOLD/FUNGI)”(1); “Other(MYSTERIOUS DISAPPEARANCE)”(1); “Other(NONPF/SURG)”(1); “Other(NONWIND)”(5); “Other (OTH-AOC)”??? (55); “Other(OTHER)”(155); “Other (OTH-LOST/MOV/STC/VMM)”(9); “Other(PFSG-oth/pf/surge)”(4) Power failure?; “Other(PLBWTR-PLB/ PLUMBING)”(10); “Other(POWER OUTAGE)”(1); “Other(PRODUCT LIABILITY CLAIM)”(1); “Other (PROPERTY DAMAGE- All others)”(1); “Other (Sewage/drain/backup/sump pump)”(31); “Other(Smoke)”(4); “Other(Smpwtr-sum)”(1); “Other(STCOTH-STC)”(2); “Other (THEFT/Auto/nonauto/onp)”(44); “Other (Unknown LCI)”(4); “Other(Vandalism/malicious mischief)”(19); “other(VEHICLE)”(32); “Other(Vehicle damage)”(2); “Other (Wat/WATER/WATER DAMAGE/ NONWEATHER)” (370) left entries that specified “weather”; “Other(WATER/PLUMBING/Appliance/Other/Overflow)”(107); “Other(Weather related-no)”(9); “Other(WTR-APL/DAM)”(119): “Other(WTR-PLB)”(353); “Other(WTR-SUM)sump pump?(86); “Other- Narrative Required (FIRE) verified not coinciding location/date of wildfire (20); “Other – Narr Req (Mysterious Disappearance/ Sewer Backup/Vandalism/Water)” (17); “Other Non-Weather Related (WATER)” (23); “Other-Physical Damage”(1); “…foyer ceiling (WATER)”(1); “Pipe Burst / Pipe Burst– Non Weather Related(WATER)” Kept “weather related” entries (80); “Place of Business(THEFT)”(1); “Plmbg – xxx (WATER)”(26); “Plug-in Electric Heater”(1); “Plumbing/Leak (Water/ weather related-no/Water damage/Water non-weather)”(241); “Power Surge”(1); “…slow leak…”(1); “Roof Leak (Weather Related-No)” (1); “Roof Leak – Poor Cond.” (8); “Roof/Exterior Leak (Weather related – no)” (2); “Settlement Related Damage”(1); “Sewage”(1); “Sewer Back-Up from Appliance/Sink/etc.”(12); “Shower Pan”(2); “Sink Related”(9); “Sink/Tub/Toilet Overflow”(1); “Smoke(SMOKE)”(2); “Smoking (FIRE)” (6); “sub Loss: (MOLD)” (1); “escape from appli (WATER)”(3); “Sub Type lost…(THEFT)”(1); “sump pump (ALL)” (41); “water pipe broke and flooded…(WATER)”(2); “Theft (ALL)” (49); “Toilet… (WATER)”(99); “Undetermined (FIRE)”(2); “Unoccupied Dwelling (Vandalism)”(1); “Vandalism/ Mischief (all)”(18); “Vehicle”(20); “Washer Hose/ Washing Machine”(40); “Water (Liability P.D./ WATER)” unclear if natural hazard related (95); “Water – Appliance/ Other/Backup/External /Internal/Other/Plumbing”(26); “water escape from washer”(3); “Water Heater”(148); “Water Leak from damaged wall”(1); “Water Mold”(2); “Water Pipes”(); “Waterbed Burst”(1); “Weatheproofing Problems”(16) 

Frozen Pipe entries left…but should we try to determine if they were during periods of “Extreme Cold”? Same with “Ice Dam” 

Left “WTR-FLD and WTR-FRZ” on assumption those are flood and freeze related; left wtr-rof assuming that was a leak from the roof and those are usually related to a hazard (hail, etc) 

Left “Rain Damage” and “Roof Leak” water claims for now….  

Left “Weather Related” (Even though they said Drain Backup…) 

Left entries related to weight of ice and snow 

 TONS of Hail entries…TONS of Wind 

Continue “Type of Loss” because all remaining “Cause of Loss” are blank: Delete “*Extended”(99); “150 – Water” deleted due to no info about whether related to weather(11); “Backup of sewer or drain” (2); “Boardup/-up”(2); “Breakage”(1); “Burglary/Theft”(3); “Collapse”(19); “Collision”(18); “Damage by Vehicle”(8); “Drain Bck”(197); “Dwelling”(2); “Ext_wdr”??(1); “Falling Object” no idea what kind or why? (21); “FIRE” checked against wildand fire date/location before deleting (872); “fire or removal” (3); “general Liability/ GL69” (6); “Glass Breakage”(4); “Hurricane”(4) checked against hurricane records and left ones that coincided with dates storms reached KY; “Insured Travel Trailer…”(1); “Liability”(8); “Mischief”(2); “Mold/Fungi”(5); “Motor Vehicles”(1); “Mysterious Disappearance”(38); “Other”(535) kept weather related; “Plumbing”(17); “Powersurge”(1); “Product Liability Claim”(3); “Property/ Damage/All others/water”(8); “Renovation”(1); “Sewage/Sewer/Backup”(63); “Smoke/Damage”(39); “Tarping”(1); “Theft/OtherThan/FromAuto/OnP” (620);  “Unknown/LCI”(234); “Vandalism/Mischief”(140); “Vehicle”(171); “Water/Damage/ApplianceRelated/non-weather”(3761) too many records to look up flood/severe cold events and compare date/ location for possibly flood/cold related…left weather-related and freezing records; “Water Non-Weather”(114); “Water Plum/bing”(10) kept water weather/flood;  “Water/Plumbing”(18); “Water-Appliance”(15); Water-Other”(70); “Water-Overflow”(10); “Water-Plumbing”(42); “Weather Related – no”(3); “Wtr-apl”(1); “Wtroth”(1); “Wtr-sum”(1);  

 27000 of 48541 original records remain after cleanup 

Impact on Agriculture

This dashboard displays the indemnity amount, which represents the dollar amount of yield loss due to wind. The dashboard also shows the number of acres that were impacted by wind. KIPDA staff included hot wind and wind/excess wind to filter for wind events. [6]

Crop losses related to wind are substantially less than residential property claims related to heat or drought. This is true for every county in the KIPDA region. It should be noted that Shelby and Trimble experienced the highest crop losses from hail. This is likely because Trimble is highly rural and agricultural, and Shelby County contains the highest agricultural acreage in the region.

  • People, who are outside or without shelter, exhibit the highest vulnerability to severe storms. Falling trees/branches  can cause injuries or in extreme cases even death.
  • Strong winds can also cause power outages. Power outages particularly affect individuals who have medical devices that require electricity.
  • Individuals that live in mobile homes are particularly vulnerable to severe storms. Mobile homes do not provide robust shelter and can be damaged or destroyed by severe storms. 
  • On average, 60 people die from wind events each year in the U.S. [7]
  • Strong winds can lead to structural damage to buildings and infrastructure.
  • Severe winds can damage trees and crops.

Hazard Vulnerability Summary Analysis

Severe storms are non-spatial hazards, and no jurisdiction is more likely to experience a severe storm than another. However, areas that are more densely populated such as Shepherdsville, Shelbyville, and La Grange are more vulnerable to property damage because these jurisdictions contain a significant built environment. 

  • Bullitt County has over $2.6 million in property damage claims due to wind annually. Property damage data is only available at the county level.
  • Bullitt County has experienced 79 high or thunderstorm wind events/21 years = 3.8 wind events annually.
  • Bullitt County has the largest population in the KIPDA region and thus has a more substantial buildings stock. The primary impact of severe storms in the region is property damage, which heightens Bullitt County’s vulnerability.

Because of these factors, Bullitt County experiences high vulnerability to severe storms. Because each of its cities contain clusters of building stock, it can be assumed that Bullitt County’s cities experience high vulnerability to severe storms as well.

  • Henry County has over $950K million in property damage claims due to wind annually. Property damage data is only available at the county level.
  • Henry County has experienced 58 high or thunderstorm wind events/21 years = 2.8 wind events annually.
  • While Henry County is significantly less developed and has substantially fewer properties than other counties, it still experiences a substantial number of property damage claims – totaling over $10 million between 2010 and 2020.

Because of these factors, Henry County experiences high vulnerability to severe thunderstorms. Because each of its cities contain clusters of building stock, it can be assumed that Henry County’s cities experience high vulnerability to severe thunderstorms as well.

  • Oldham County has over $3.1 million in property damage claims due to wind annually. Property damage data is only available at the county level.
  • Oldham County has experienced 77 high or thunderstorm wind events/21 years = 3.7 wind events annually
  • Oldham County has the second largest population in the KIPDA region and thus has a more substantial buildings stock. The primary impact of severe thunderstorms in the region is property damage, which heightens Oldham County’s vulnerability.

Because of these factors, Oldham County experiences high vulnerability to severe thunderstorms. Because each of its cities contain clusters of building stock, it can be assumed that Oldham County’s cities experience high vulnerability to severe thunderstorms as well.

  • Shelby County has approximately $1.6 million in property damage claims due to wind annually. Property damage data is only available at the county level.
  • Shelby County has experienced 84 high or thunderstorm wind events/21 years = 4 wind events annually.
  • Shelby County has the third largest population in the KIPDA region and thus has a more substantial buildings stock. The primary impact of severe thunderstorms in the region is property damage, which heightens Shelby County’s vulnerability.

Because of these factors, Shelby County experiences high vulnerability to severe thunderstorm. Because each of its cities contain clusters of building stock, it can be assumed that Shelby County’s cities experience high vulnerability to severe thunderstorms as well.

  • Spencer County has over $500K  in property damage claims due wind annually. Property damage data is only available at the county level.
  • Spencer County has experienced 45 high or thunderstorm wind events/21 years = 2.1 wind events annually. It is tied with Trimble County for the fewest number of events.
  • While Spencer County is significantly less developed and has substantially fewer properties than other counties, it still experiences a significant number of property damage claims – over $6 million between 2010 and 2020.

Because of these factors, Spencer County experiences high vulnerability to severe storms. Because Taylorsville contain clusters of building stock, it can be assumed that Spencer County’s cities experience high vulnerability to severe storms as well.

  • Trimble County has over $345K in property damage claims due wind annually. Property damage data is only available at the county level.
  • Trimble County has experienced 45 high or thunderstorm wind events/21 years = 2.1 wind events annually. It is tied with Spencer County for the fewest number of events.
  • Trimble County experienced less than $3 million in property damage claims between 2010 and 2020, which is substantially less than any other county.
  • Trimble County is largely undeveloped and contains a limited building stock compared to other KIPDA counties.

Because of fewer property claims and limited development, Trimble County has a high vulnerability to severe storms. Because Trimble County’s cities contain clusters of building stock, it can be assumed that Trimble County’s cities experience high vulnerability to severe storms as well.

References

[1] National Oceanic and Atmospheric Administration (2021). Storm Events Database. [Data set]. National Centers for Environmental Information. Retrieved from https://www.ncdc.noaa.gov/stormevents/

[2] Allen, J. T. (2018). Climate Change and Severe Thunderstorms. Oxford Research Encyclopedia of Climate Science. doi:10.1093/acrefore/ 9780190228620.013.62

[3] Wuebbles, D. J., Kunkel, K., Wehner, M., & Zobel, Z. (2014). Severe Weather in United States Under a Changing Climate. Eos, Transactions American Geophysical Union, 95(18), 149-150. doi:10.1002/2014eo180001

[4] Molteni, M. (n.d.). Climate Change Is Bringing Epic Flooding to the Midwest. Retrieved July 24, 2020, from https://www.wired.com/story/for-the-midwestepic-\looding-is-the-face-of-climate-change/

[5] Velzer, R. (2019, February 07). Ky. Won’t Escape Human-Driven Climate Change, Federal Report Warns. Retrieved July 24, 2020, from https://wfpl.org/kentucky-wont-escape-human-driven-climate-change-federal-report-warns/

[6] USDA Risk Management Agency (2021). Cause of loss. [Data set]. Retrieved from https://www.rma.usda.gov/en/Information-Tools/Summary-of-Business/Cause-of-Loss

[7] National Weather Service (2021). Weather Related Fatalities 2020. National Weather Service. Retrieved from https://www.weather.gov/hazstat/