Project 5: Feasibility Study for In-Home Storm Shelters in Florida Residential - PowerPoint Presentation

1. Project 5: Feasibility Study for In-Home Storm Shelters in Florida Residential HomesPresented to the Florida Building Commission Building (Structural) Technical Advisory CommitteeState of Florida Department of Business and Professional RegulationJune 24th, 2014David O. Prevatt, Ph.D., P.E. (MA), Associate ProfessorAustin Thompson, Keith Clabaugh, David B. Roueche Graduate Assistants

2. Purpose of Study Develop guidelines for strengthening a room or an area within an existing home necessary to achieve an acceptable level of protection from a severe windstorm. Is the existing shelter criteria appropriate for Florida? Should FL shelters be designed to resist tornado loads? What are the cost/benefit implications of shelters in homes?Outline a test program (if needed) for Florida shelters.

3. What minimum requirements for structural strength and affordability are needed to provide safety to life and property from wind (hurricane and tornado) hazards for existing residential buildings?

4. Hurricanes per CountyFLORIDA’S STORM SURGE RISKStorm surge heights ranges 6-9 ft : Zone 1 28-42 ft : Zone 5

5. Wind Damage Statistics, 1950-2013Hurricanes31 of 91 US landfalling hurricanes hit Florida’s coasts (34%)Only two design level wind events (Andrew and Charley)$58 B in losses, 73 dead & 1,253 injured (1992 & 2004)Tornadoes3,183 tornadoes recorded in Florida (50 per year)Only 39 EF-3 and larger tornadoes, 319 EF-2 tornadoes1 in 5 Florida tornadoes are nocturnal (late evening to night)45.8% of the killer tornados in Florida were nocturnal events

6. Extreme Wind Speed Distribution Tornado Distributions Population DensityASCE 7 – Category II Design Wind Speed Map

7. The Florida 2004 Hurricane Season4 major hurricanes47 dead, $45 B in damages, 40,000 homes damaged2.8 million persons evacuated before Frances1.7 million sought temporary shelter

8. The April 1966 Tampa Tornado FamilyF4 -100 mile path,11 fatalities, 3,350 injuries1 school, 280 homes & 150 trailers demolishedWidespread looting in 2 counties$50 million estimated damage

9. Tornado Damage Patterns vs Hurricane Damage Damaged masonry houseKissimmee, FL during 1998 F3 Wood-framed structure destroyed by EF2 tornado Lake Placid 2012Window broken  Roof removed  gable wall topples  exterior (side wall) falls  interior partitions collapse  occupants injured  fatalities occur

10. Aged Residential Buildings (pre-Andrew codes)-9 million single-family housing units in Florida-6.3 million homes are 20 years old & older-Concentrated in North Central Florida and SEi.e. 70% of our homes do not have wind resistant details – vulnerable to hurricanes

11. 25-28 April 2011 Tornado Outbreaks 44% of the bodies recovered inside private homes47% of the injuries were inside single-family homes Exterior wall collapsedRoof structure fails, wall topple

12. Wind Related Mortality and Morbidity RatesThe degree of building failure is highly correlated with risk of death or injurySevere damage of buildings was most significantly associated with PTSD, and mental depression outcomesMost major injured had not covered themselves with blankets, or pillows

13. DefinitionsStorm Shelter: a structure constructed in accordance with NSSA/ICC-500, designated for use during a severe wind storm event (hurricanes and tornadoes)Residential storm shelter: serving occupants of dwelling units, occupant load up to 16 persons.Host Building: A building not designed or constructed as a storm shelter that totally/partially encloses a storm shelter.

14. Storm SheltersSmall interior rooms can survive direct tornado hitKiesling and Goolsby (1974) a reinforced small room to resist tornado loads and provide life safetyFeasible in concrete or wood material but must not rely upon host building for supportShelter survives despite damage to the host building Door strength and locking details are critical

15. Design Criteria for Storm SheltersNSSA/ICC-500 (Tornado)100 psf roof live load200 mph design windspeed (FL)15 lb 2x4 traveling at 100 mphResist collapsed host building load Independent of host buildingProvision of ventilation3 sf space per personUF recommends no changeMissile impact testing of storm shelter at Texas Tech University

16. Hurricane Shelter Design CriteriaNSSA / ICC-500Design wind speed (see map) Roof live load = 50 psfMissile Test 9 lb 2x4 @ 0.4 times wind speedMin. fl elev. 1 ft above floodOccupant density = 7 sfVenting = 4 in2 per occupantSanitation = no, water = no, Flashlights = yesDoors and windows = 1.2 times design pressureFor Florida: Combination Shelter Recommended – use more stringent criteria

17. Wood-framed Shelter Wall Systems9 lb 2x4 missile at 25mph15 lb 2x4 missile at 100 mphDozens of wall configurations tested and impact resistance established.There is some potential to reduce impact velocity of missile for Florida but economic justification is not there.Lin et al. (2007) paper suggests missiles may travel faster than 100 mphStructural framing shown generically – must be designed for wind pressureTTU, 1998FPL, 2013Clemson, 2000

18. Storm Shelters and EvacuationPeople have less risk of injury in homes than in carsNorthbound volume on the Turnpike was 14 times the normal during Frances evac.Reducing congestion traffic will benefit persons most at risk (leaving storm surge)Recovery and repair occurs faster when occupants shelter-in-placeEvacuation traffic from Pinellas County before Hurricane Charley. (Source: Sun Times)

19. Shelters Protect PeopleEF 5EF 5EF 4

20. Wall Corner Failure Joplin, MO

21. Interior Walls AnchoredInterior Walls not Anchored

22. Interior Wall Anchors – zero! available capacity to support walls without roof, (Joplin, MO)

23. Host Building Structural Failure2014 research at UWO, Canada studied stability of walls once roof is lostAttached cables and load spreaders to walls to examine failure of exterior wall corner and Tee-joint connectionsThree Little Pigs Test Building

24. Interior Partition to Exterior Wall Strength

25. Failure Progression of Exterior WallsStedman (2014)Interior-to-exterior wall connection strength directly affects overall capacity of exterior wall. BEFORE TESTJOINT D10 IS FAILINGBOTH JOINTS FAILED

26. Corner return provides hold-down anchor to building. Preventing rotation increases wall stabilityAngle increasesCorner Strength and Wall Stability

27. Recap: Why Florida Specific Shelters?No basements for protection in our Florida housesStorm surge threat - coastal population MUST evacuateNocturnal tornados pose a higher risk of injury/deathHurricane protection must last for more than 2 hoursOlder homes features enable disproportionate collapse

28. Protect the People – Inside the Shell (The Egg)Safeguard the Structure from Damage (The Carton)

29. Egg and Carton Design ComponentsUse structural fasteners and connections Ring shank nails, metal hurricane straps More anchor bolts into foundationContinuous structural sheathing on wallsReinforce exterior building corners to hold wallsUtilize strength of interior partition wall buttressesUse same anchor and metal tie connections as exterior wallsBrace exterior walls using strengthened interior wallsShelter Design (per NSSA / ICC-500, combination shelter)200 mph for wind pressure calculations100 mph for missile impact design speedSelect interior room with no exterior walls

30.

31. Corner RetrofitTee-Joint RetrofitFEMA 320 RecommendedStorm Shelter Locations

32. Shelter C (Laundry Room)Shelter spaces may be too small for long-term (12-24 hour) sheltering.More feasible renovation to create space:GIVE AN ORDINARY ROOM AN EXTRAORDINARY PURPOSE (FLASH, 2010), SERRI Report# 89960-01

33. PHASE IIEnhanced robustness of residential (host) buildings is achievable by providing inexpensive shear connectors at wall corners and at Tee-joint connections between exterior and interior walls. California Corner (APA Advanced Framing Guide)Conventional Corner (IRC 2012)

34. Phase 2 TestsStrength of wood-frame wall-to-wall connectionsRelevant Sections of Florida Building CodeSection R602 Wood Wall Framing (ICC2009)Table 602.3(1) Fastener Schedules for Structural MembersItem 7: Built-up corner studs: 10d at 24 in. o.c.Item 16: Stud-to-sole plate:Item 17: Top plate to stud:Item 18: Top plate lap at corner & intersectionsEngineering Justification:Corner framing strength and details Interior-to-Exterior wall connection strengths

35. Scope of WorkIdentify construction details for corner and Tee-joint construction in existing homes (say pre-1994 homes)Construct and evaluate wind resistance per ASTM E330corner tests interior-to-exterior testsConstruct, retrofit & evaluate wind resistance of retrofitsretrofitted corner tests retrofitted interior-to-exterior testsEvaluate cost to implement in existing house (actual retrofit?)Produce structural design calculations example If indicated, develop code language for specifying wall-to-wall connections in wood-frame Chap. R602.

36. Plan View of Test Setup

37. Test 1: Corner strength with roof in placeHAPLA

38. HAPLATest 2: Corner strength after roof failure

39. HAPLATest 3: Interior-exterior wall strength

40. AcknowledgementsThe authors gratefully acknowledge the National Science Foundation for partial financial support for the tornado climatology item of this study. That work was performed under research grant 1150975. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.Leslie Chapman – Federal Alliance for Safe HomesDr. Ernst Kiesling – National Storm Shelter AssociationRandy Shackelford – Simpson Strong-TieDr. Gregg Kopp – University of Western Ontario

41. Thank you!Questions/Comments

42. Thank you!Questions/Comments

43. Shelter-in-place OptionsInstall/Construct a FEMA shelterProvide near-absolute protection Expect damage to house and long recovery timeRetrofit/Strengthen the building envelope and structural framingReduce property damage, yet it may be insufficient to ensure life safety. Retrofit/Strengthen the home to near-absolute protection. Risk may not warrant the costs for such renovation.Retrofit/Strengthen the vulnerable connections and build a hardened interior room.Reduces damage to the house and protects the lives of the peopleEgg and Carton Approach

44. Impact Resistance ResearchTTU Carter MS 1998Tested 30 wood and 3 masonry composite wall systemsClemson 200036 wood-framed wall system Developed linear momentum model related to ultimate strength of a FEMA/ICC Storm ShelterFPL 2013Examined the effectiveness of 30 wood panel wall designs for use in retrofitting existing building with storm shelters. Missile impact testing at Forest Product Laboratories

45.

46.

47. http://csc.noaa.gov/hurricane_county_strikes/index.html?county=pinellas&state=florida

48. Traffic volumes on major Florida roadways just prior to Hurricane Frances, 2004 (FDOT, 2009)

49. U.S. Catastrophe Losses, 1993-2012Tornadoes: 36% or $141 BillionSource: ISO’s Property Claims Services Unit49Source: ISO’s Property Claims Services

50. Annual Probability of Wind Speed Exceedance due to Tornadoes (Note: ASCE 7 Windspeed POE is based on geographic location (i.e. contour lines) and the tornado POE are for the entire stateTwisdale and Dunn (1983) and Banik et al. (2007)Methodology for prediction of tornado wind probabilities

51.

52. Gurley and Masters (2011) “Post-2004 hurricane field survey of residential building performance,” Nat. Haz. ReviewImproved Building Codes = Fewer Wind Damaged HousesYear House Built