Storm Shelter Design: Overview of IBC and ICC-500 Requirements Including Safe Rooms and Significant Changes in ICC 500-2020

Home , Safe room

University of Minnesota Structural Engineering Webinar 1/26/2021

Marc L. Levitan, Ph.D.

Lead Research Engineer National Windstorm Impact Reduction Program National Institute of Standards and Technology

Chair, Storm Shelter Standards Committee: ICC 500-2008 Chair, Structures/Chapter 3 Work Group: ICC 500-2014 Chair, Storm Shelter Standards Committee: ICC 500-2020 Chair, ASCE 7-22 Tornado Task Committee Co-Chair, ASCE/AMS Std on Wind Speed Estimation in Tornadoes Cmte Acknowledgements

▪ International Code Council – ICC 500 Committee Secretariat Staff Dave Bowman (2008 and 2014 eds) and Kim Paarlberg (2020 ed.) – ICC 500 committee members and interested parties ▪ FEMA, Building Science & Hazard Mitigation – Slides with FEMA logo at bottom used with permission ▪ National Storm Shelter Association ▪ NIST ▪ Many collaborators over the years

Source: Marc Levitan / LSU Hurricane Center

Design of Storm Shelters | 2 | Presentation Outline

▪ Shelter/Safe Room Terminology and Purpose ▪ Shelter Design Considerations ▪ Development of the Shelter Standard ▪ Structural Provisions, incl. Wind Loads ▪ Pressure and Impact Testing Primary ▪ Model Building Code and State Requirements Focus on for Storm Shelters Tornado Shelters ▪ FEMA Safe Room Criteria ▪ FEMA Mitigation Grants (help pay for safe rooms!) ▪ Key Changes in ICC 500-2020 ▪ ASCE 7-22: Tornado Loads for Conventional Buildings ▪ Additional Resources

Design of Storm Shelters | 3 | Terminology Stand-alone Shelter Public Shelter Red Cross In-residence Shelter Safe Room shelter Community Shelter Storm Refuge Area Shelter Severe Emergency Weather Shelter Community Safe Shelter Area Room Residential Best Available Shelter-in- Shelter Refuge Area place

Design of Storm Shelters | 4 | Terminology

▪ STORM SHELTER. A building, structure or portion thereof, constructed in accordance with ICC 500, designated for use during a tornadoes, hurricanes, and other severe

windstorms. (ICC 500-2020)

▪ SAFE ROOM. An interior room, space within a building, or entirely separate building, designed and constructed to provide near absolute life-safety protection for its occupants from tornadoes or hurricanes. Safe rooms are designed and constructed to meet the criteria in FEMA P-

361 or FEMA P-320. (FEMA P-361, 3rd ed.)

Design of Storm Shelters | 5 | Terminology

▪ BEST AVAILABLE REFUGE AREA. Building area (or areas) that has been determined by an RDP to be least vulnerable to the life threatening effects of extreme-wind incidents relative to other building areas. These areas were not specifically designed as tornado safe rooms, and as a result, occupants may be injured or killed during a Tornado Protection: tornado. However, people in best Selecting Refuge available refuge areas are less likely to Areas in Buildings, FEMA P-431, Second be injured or killed than people in other Edition / October 2009 areas of a building

Design of Storm Shelters | 6 | ICC 500: Purpose

The purpose of this standard is to establish minimum requirements to safeguard the public health, safety and general welfare relative to the design, construction and installation of storm shelters constructed for protection from tornadoes, hurricanes and other severe windstorms.

This standard is intended for adoption by government agencies and organizations for use in conjunction with applicable codes to achieve uniformity in the technical design and construction of storm shelters.

ICC 500-2020

Design of Storm Shelters | 7 | FEMA 361: Purpose

The purpose and scope of FEMA P-361 is to provide guidance—including emergency management considerations – for safe rooms that provide near- absolute protection.

Near-absolute protection: Based on our current knowledge of tornadoes and hurricanes, the occupants of a safe room built according to FEMA P-361 will have a very high probability of being protected from injury or death.

FEMA P-361, 3rd ed.

Design of Storm Shelters | 8 | Hattie A. Watts Elementary This Standard Hopes to Prevent This... School Patterson LA Hurricane Andrew, 1992

Source: Marc Levitan / LSU Hurricane Center

Design of Storm Shelters | 9 | And This…

Enterprise High School Enterprise AL March 1 2007 Tornado

Source: FEMA

Design of Storm Shelters | 10 | School Shelter And This… North Carolina Hurricane Fran, 1995

Source: NOAA

Design of Storm Shelters | 11 | And This…

Source: FEMA

Turner Agricivic center Arcadia FL Hurricane Charley, 2004

Source: Marc Levitan / LSU Hurricane Center

Design of Storm Shelters | 12 | Lincoln High School McCLellenville SC And This… Hurricane Hugo, 1989

Source: NOAA/National Weather Service

Design of Storm Shelters | 13 | And This…

Plaza Towers Elementary School Moore OK May 20, 2013 Tornado

Source: Marc Levitan / NIST

Design of Storm Shelters | 14 | Plaza Towers Elementary School Moore OK And This… May 20, 2013 Tornado

https://doi.org/10.6028/NIST.sp.1164

Design of Storm Shelters | 15 | Shelter and Safe Room Purpose

In-Residence Safe Room Winston County Commission Joplin, MO, May 22, 2011 Community Safe Room Arley, AL, November 30, 2016

Source: FEMA Source: FEMA Success from Joplin, MO Tornado

Residential Safe Room ▪ Installed in garage ▪ Passed Texas Tech testing ▪ 2 occupants survived ▪ Testimonial video: http://www.fema.gov/media- library/assets/videos/81425

FEMA P-361 1-Day Training - Introduction and Module A1 17 Success from Joplin, MO Tornado

FEMA P-361 1-Day Training - Introduction and Module A1 18 Track Record of Success

“FEMA publications presenting design and construction guidance for both residential and community safe rooms have been available since 1998. Since that time, thousands of safe rooms have been built, and a growing number of these safe rooms have already saved lives in actual events. There has not been a single reported failure of a safe room constructed to FEMA criteria.” Source: FEMA P-361, 3rd ed

Similarly, failures of tornado shelters constructed to the ICC 500 standard (first published in 2008) have not been reported. ICC 500: Scope

This standard applies to the design, construction, installation and inspection of storm shelters constructed for the purpose of providing protection from tornadoes, hurricanes and other severe windstorms.

Storm shelters shall be constructed as either separate, detached buildings or rooms or spaces within new or existing buildings.

Design of facilities for use as emergency shelters after the storm is outside the scope of this standard.

ICC 500-2020

Design of Storm Shelters | 20 | Presentation Outline

Design of Storm Shelters | 21 | Differences Between Tornado and Hurricane Shelter Design Considerations

▪ Numerous and Significant – Hazards – Warning Time – Duration of Stay

Design of Storm Shelters | 22 | Differences Between Tornado and Hurricane Shelter Design Considerations Primary Hazards

Tornado Hurricane – Extreme Winds – Extreme Winds and Tornadoes – Windborne Debris – Windborne Debris – Laydown, Falling Debris, and – Laydown, Falling Debris, and Collapse Hazards Collapse Hazards – Atmospheric Pressure Change – Extreme Rain (roof load) – Flash Flooding – Storm Surge Flooding and – Lightning Waves – Inland Flooding – Floodborne Debris – Lightning

Follow-on Hazards

– Fire following damage – Shelter exits blocked by debris

Design of Storm Shelters | 23 | Differences Between Tornado and Hurricane Shelter Design Considerations

Warning Time Tornado Hurricane – Minutes – Days

Related Considerations Community Shelters Tornado Hurricane – Unlocking of shelter doors – Travel to shelter – Shelter in place or adjacent – Accessibility of shelter from building transportation network – Rate of ingress to shelter space – Parking – Signage to identify shelter location – Backup power, water, sanitation Design of Storm Shelters | 24 | Differences Between Tornado and Hurricane Shelter Design Considerations

Duration of Stay

Tornado Hurricane – Minutes to Hours – Hours to Days

Related Considerations Community Shelters Tornado Hurricane – Space required per person – Space required per person on the order of 5 sq ft on the order of 20 sq ft – Managed facilities – Utilities (water, electricity, food handling)

Design of Storm Shelters | 25 | Presentation Outline

Design of Storm Shelters | 26 | Storm Shelter Standard

ICC/NSSA Standard for the Design and Construction of Storm Shelters ▪ Initial development began in Spring 2003 as a collaborative effort between – International Code Council – National Storm Shelter Association – Federal Emergency Management Agency ▪ 1st edition: 2008

▪ 2nd edition: 2014 https://codes.iccsafe.org/content/ICC5002014 ▪ 2nd edition + Commentary: 2016 rd ▪ 3 edition: 2020 https://codes.iccsafe.org/content/ICC5002020P1 ▪ 3rd edition + Commentary: early 2021

Design of Storm Shelters | 27 | ICC Standard Consensus Process

▪ Form ‘balanced’ committee ▪ Gather input from scientific, technical, and user communities ▪ Develop draft standard ▪ Publish draft standard for public comment ▪ Formal, public committee meeting to address all comments ▪ ANSI approval ▪ Publish standard ▪ Available for incorporation in building codes ▪ Available for direct adoption by states or communities

Design of Storm Shelters | 28 | 2008 Shelter Standard Committee Members

▪ Balanced Committee – General Interest, Producer Interest, User Interest ▪ Committee Makeup – Code Officials – Emergency Managers and Planners • Federal (FEMA) and State Officials – Industry • NAHB, NSSA, PCA, NCMA, AISI, Ingersoll Rand, Remagen Safe Rooms – University faculty members: • LSU, Texas Tech – Consultants and Users • URS, JDH Consulting, PBA Architects, IBHS ▪ Included a number of structural engineers ▪ And several wind engineers – John Holmes, Ernst Kiesling, Marc Levitan, Tim Reinhold ▪ Numerous ‘Friends of the Committee’ – Particularly from industry, testing labs, and FEMA

Design of Storm Shelters | 29 | Primary Source Documents/References

▪ NSSA Industry Standard – Standard for the Design, Construction, and Performance of Storm Shelters – primarily addresses residential shelters ▪ FEMA 320: Taking Shelter from the Storm: Building a Safe Room Inside Your House – in-residence shelters ▪ FEMA 361: Design and Construction of Community Storm Shelters – standalone or as part of a larger building ▪ ARC4496: Guidelines for Hurricane Evacuation Shelter Selection – addresses community shelters – selection of existing buildings ▪ Florida public shelter design criteria – known as Enhanced Hurricane Protection Area (EHPA) criteria – section 423.25, Florida Building Code

Design of Storm Shelters | 30 | ICC 500: Scope

▪ Hurricane and Tornado Shelters ▪ Residential and Community Shelters ▪ Includes Considerations for – Architectural – Structural – Mechanical – Electrical – Plumbing – Other

Design of Storm Shelters | 31 | ICC 500: Chapters

▪ Chapter 1 – Application and Administration ▪ Chapter 2 – Definitions ▪ Chapter 3 – Structural Design Criteria ▪ Chapter 4 – Siting ▪ Chapter 5 – Occupancy, Means of Egress, and Accessibility ▪ Chapter 6 – Fire Safety ▪ Chapter 7 – Shelter Essential Features and Accessories ▪ Chapter 8 – Test Methods for Impact and Pressure Testing ▪ Chapter 9 – Referenced Standards ▪ Appendix A - Storm Shelter Preparedness and Emergency Operations Plan (new in ICC 500-2020)

Design of Storm Shelters | 32 | Presentation Outline

Design of Storm Shelters | 33 | ICC 500-2014 Chapter 3 Structural Design Criteria

▪ Section 301: General ▪ Section 302: Load Combinations ▪ Section 303: Loads ▪ Section 304: Wind Loads ▪ Section 305: Debris Hazards ▪ Section 306: Component Design and Testing ▪ Section 307: Weather Protection ▪ Section 308: Connection to Foundations or Slabs ▪ Section 309: Penetrations of Envelope by Systems and Utilities

Design of Storm Shelters | 34 | General

301.1 Scope. Loads and load combinations shall be determined in accordance with ASCE 7 unless otherwise noted. Structural elements of the storm shelter shall be designed in accordance with the appropriate material design standard specified in the applicable building code to sustain the loads prescribed in ASCE 7, as modified by this chapter, and combined in accordance with the load combinations of ASCE 7, as modified by Section 302.

▪ Loads based on the ASCE 7-10 Standard, Minimum Design Loads and Associated Criteria for Buildings and Other Structures, with some modifications, including – Wind Loads • Increased wind speeds • Changes to various wind load parameters • Incorporation of effects of Atmospheric Pressure Change in tornadoes – Increased roof live load to account for debris – Increased rainfall rates for design against ponding on the roof (and for drainage) – Increased design flood elevations, and associated minimum floor elevations

Design of Storm Shelters | 35 | Load Combinations

302.1 Strength design. For strength design or load and resistance factor design (LRFD), use the load combinations stated in ASCE 7, Section 2.3 with W determined in accordance with Section 304 of this standard. Exception 1 to ASCE 7 Section 2.3.2 shall not apply. 302.2 Allowable stress design. For allowable stress design (ASD), use the load combinations stated in ASCE 7, Section 2.4 with W determined in accordance with Section 304 of this standard.

▪ ASCE 7 Exception 1 allows reduction in Live Load from 1.0L to 0.5L for some combinations ▪ Load Combination provisions of ICC 500-2014 (and -2008) have significant omissions – as ICC 500 changes more than just W – This issue was identified in the Commentary, which was published later ▪ Other revised loads that must also be included in ASCE 7 combos: – Rain loads for hurricane shelters – Roof live loads – Hydrostatic loads – Flood loads

Design of Storm Shelters | 36 | Rain Loads

303.1 Rain loads. Rain loads shall be determined in accordance with ASCE 7. Rainfall rates for hurricane shelter roofs shall meet the following requirements: 303.1.1. Rainfall rate. The rainfall rate shall be determined by adding 6 inches (152.4 mm) of rainfall per hour to the rainfall rate established from Figure 303.2.

▪ Rainfall rate maps are from the 2012 International Plumbing Code – 100 year Mean Recurrence Interval (MRI)

▪ The addition of 6 inches/hour to the mapped rainfall rate is an approximation of the 10,000-year MRI

▪ Debris and damage to the shelter roof cover may block primary and/or secondary drains.

Design of Storm Shelters | 37 | Roof Live Loads

303.2 Roof live loads. Storm shelter roofs shall be designed for minimum live loads specified in ASCE 7, but not less than the following: Tornado shelters: 100 pounds per square foot (4.8 kN/m2) Hurricane shelters: 50 pounds per square foot (2.4 kN/m2)

▪ Provide for overall robustness to account for loads due to nonspecific debris hazards ▪ Specifically identified laydown and collapse hazards addressed in Section 305.3

Source: Long Phan / NIST

Design of Storm Shelters | 38 | Hydrostatic Loads

303.3 Hydrostatic loads. Underground portions of storm shelters shall be designed for buoyancy forces and hydrostatic loads assuming that the ground water level is at the surface of the ground at the entrance to the storm shelter, unless adequate drainage is available to justify designing for a lower ground water level.

▪ Buoyancy failures of below ground residential shelters have been observed on multiple occasions

Design of Storm Shelters | 39 | Flood Loads

303.4 Flood loads. Flood loads shall be determined in accordance with ASCE 7. The design flood elevation shall equal the minimum floor elevation as specified in Section 401 of this standard.

401.1.1 Minimum floor elevation of community shelters. The lowest floor used for the occupied shelter and occupant support areas of a community shelter shall be elevated to or above the higher of the elevations determined by: 1. The flood elevation, including coastal wave effects, having an 0.2-percent annual chance of being equaled or exceeded in any given year; or 2. The flood elevation corresponding to the highest recorded flood elevation if a flood hazard study has not been conducted for the area; or 3. The maximum flood elevation associated with any modeled hurricane category, including coastal wave effects; or 4. The minimum elevation of the lowest floor required by the authority having jurisdiction for the location where the shelter is installed; or 5. Two feet (610 mm) above the flood elevation having a 1 percent annual chance of being equaled or

exceeded in any given year. Tornado Shelters Tornado Exception: Items 1 and 3 shall not apply to shelters designed, constructed, designated and used only as tornado shelters.

Design of Storm Shelters | 40 | Wind Loads

304.1 General. Wind loads shall be determined using ASCE 7, except as modified by this section.

▪ Wind pressures and forces determined using ASCE 7-10 procedures and equations, with modifications to

Kz = Velocity Pressure Exposure Coefficient

Kzt =Topographic Factor

Kd = Wind Directionality Factor V = Shelter Design Wind Speed Enclosure Classification Internal Pressure Coefficient

Design of Storm Shelters | 41 | Velocity Pressure per ASCE 7-10

▪ Velocity Pressure (psf): 2 qz = 0.00256KzKztKdV

Kz = Velocity Pressure Exposure Coefficient

Kzt =Topographic Factor

Kd = Wind Directionality Factor V = Design Wind Speed (mph)

Design of Storm Shelters | 42 | Velocity Pressure per ICC 500-2014

▪ Velocity Pressure (psf): 2 qz = 0.00256KzKztKdV

Kz = Velocity Pressure Exposure Coefficient

Kzt =Topographic Factor

Kd = Wind Directionality Factor V = Shelter Design Wind Speed Items in red are modified from ASCE 7

Design of Storm Shelters | 43 | Kz Velocity Pressure Exposure Coefficient

Velocity profiles remain unchanged, but selection of exposure categories is modified

▪ Exposure C (open terrain) required for both: – Main Wind Force Resisting System (MWFRS), and – Components and Cladding (C&C)

▪ Exception for Hurricane Shelters – MWFRS only – Exposure B permitted when this exposure prevails in all wind directions

Design of Storm Shelters | 44 | ‘Type’ of Exposure B Important

Roughness unlikely to change no matter how intense the damage

Roughness could change from B to C during a severe hurricane

Design of Storm Shelters | 45 | Example of Pre-storm Exposure B, Becoming Exposure C During Storm

Hurricane Andrew

Design of Storm Shelters | 46 | Kzt Topographic Factor

Accounts for speedup in wind near top of hills, ridges and escarpments

▪ Tornado Shelter – Kzt = 1.0 – Effects of topography on tornado wind speeds are unknown

▪ Hurricane Shelter – Per ASCE 7

Design of Storm Shelters | 47 | Kd Directionality Factor

Kd Accounts for reduced probability of maximum wind speed occurring at the wind direction that maximizes pressure coefficients

▪ ASCE 7 – Kd = 0.85 ▪ ICC 500 – Kd = 1.0 – Changing wind direction may bring maximum or near maximum wind speeds over a wide range of wind directions

Design of Storm Shelters | 48 | V Shelter Design Wind Speed

Ultimate Wind Speed 3-sec gust, 10 m, open terrain

Tornado Shelter – Based on FEMA 361 – 4 zones across the country – Developed from deterministic analysis – Annual probability of exceedance estimated to vary from 0.5 x 10-4 to 10-6 /year, i.e., • MRI ranges from approximately 20,000-1,000,000 years – Based on best available data at the time

Design of Storm Shelters | 49 | Tornado Shelter Design Wind Speed

Design of Storm Shelters | 50 | Tornado Shelter Design Wind Speed

Errors in ICC 500-2008 and -2014 These speeds are for HURRICANE shelters

Design of Storm Shelters | 51 | Shelter Design Wind Speed

Hurricane Shelter – Annual probability of exceedance 0.01%, i.e., • 10-4 / year, or • MRI = 10,000 years, or • 0.5% chance of exceedance in 50 years – Map developed by ARA using hurricane simulation model and data similar to hurricane isotachs ASCE 7-10 wind maps – Minimum wind speed is 160 mph • Rationale: protection from hurricane-generated tornadoes • Consistent with minimum tornado design speed along hurricane coast – 2020 hurricane map created using same updated climatology and methodology in ASCE 7-22 basic wind speed maps

Design of Storm Shelters | 52 | Hurricane Shelter Design Wind Speed

Design of Storm Shelters | 53 | ASCE 7-10 Risk Category III-IV Speeds

Hurricane Shelter Design Wind Speeds ≈ 10-30 mph > ASCE 7 along the Gulf and Atlantic Coastlines

Design of Storm Shelters | 54 | Enclosure Classification

▪ ICC 500 Section 304.6 ▪ Use procedure from ASCE 7 to determine classification, except for Community Shelters – the largest door or window or other protected opening on each side of the building must be considered in turn as being open

Experience has shown that even if openings are not breached, people may open shelter doors and windows during the storm

Design of Storm Shelters | 55 | Opening Protective Installation

▪ Tornado Shelters – Must be permanently affixed – Must be manually operable from inside the shelter ▪ Hurricane Shelters – No special requirements

Warning: Designing impact protectives for hurricane shelters that require storage and installation before a storm, or special training, access, keys, tools, or other actions should be carefully considered.

Any type of operable protective device should have a regular maintenance and testing program.

Design of Storm Shelters | 56 | Performance Observations: Door Systems

▪ 2004 Hurricanes Charley and Ivan in Florida ▪ Shelter doors will be opened during the storm, for numerous reasons including – Late arrivals – Investigating damage – Moving between shelter areas – People wanting to smoke – Etc. - human nature

Design of Storm Shelters | 57 | Performance Observations: Window Protection Systems

Hurricane Ivan Sherwood Elementary School, Pensacola, FL

Design of Storm Shelters | 58 | Performance Observations: Window Protection Systems

Hurricane Ivan West Florida High School of Advanced Technology

Design of Storm Shelters | 59 | Performance Observations: Window Protection Systems

▪ 2004 Hurricanes Charley and Ivan in Florida ▪ While most shutters seemed to have performed well, accordion shutters were observed to fail in several ways – Shutters ripped off tracks – Shutters not closed and latched properly before storm – Shutters popped open during storm because of failed latches or locks

Design of Storm Shelters | 60 | Window Protection Systems Operations, Maintenance, and Training Issues

Bottom line - will protection really be there and provide designed protection Tampa-area EHPA Fixed debris impact protection Shelter systems avoid this problem

Design of Storm Shelters | 61 | Tornado Shelter Window Protection

▪ Impact resistant glazing ▪ Permanently affixed external protection (e.g. perforated metal screens) ▪ Interior shutters

Interior operated safe room shutters in multi-purpose classroom/safe room. Image on left is normal usage; image on right shows shutters in ‘lock down’ position where they are closed and latched (source: FEMA P-361 3rd ed)

Design of Storm Shelters | 62 | Internal Pressure

APC = Atmospheric Pressure Change

▪ Tornado shelters – APC must be considered – 3 options

– GCpi = +/- 0.18 + APC contribution, or – GCpi = +/- 0.18 + design for venting, or – GCpi = +/- 0.55 ▪ Hurricane shelters – No APC (atmospheric pressure changes slowly)

– GCpi per ASCE 7

Design of Storm Shelters | 63 | Shelters in Host Buildings

304.8 Shielding of storm shelters by host and adjacent buildings. Storm shelters enclosed in, partially enclosed in or adjacent to host buildings or adjacent to other buildings not designed for the load requirements of Chapter 3 shall be designed considering the host building and adjacent buildings to be destroyed and the shelter to be fully exposed.

▪ Shelters Enclosed or Partially Enclosed in a Host Building – e.g., in-residence shelter, shelter is wing of a building – Wind loads must consider host building has been destroyed and the shelter is completely exposed – Shelters must be designed to resist the maximum force that could be transmitted to the storm shelter equal to the ultimate failure strength of the host building connection or element being connected

Design of Storm Shelters | 64 | Shelters Connected to Host Buildings

304.9 Storm shelters connected to host buildings. Where an element or component of the host building is connected to a storm shelter, the storm shelter shall be designed to resist the maximum force that could be transmitted to the shelter equal to the ultimate failure strength of the connection or element being connected, whichever is lower, concurrent with the other wind loads on the storm shelter required by Chapter 3.

Shelter Front Shelter Front Wall Wall

Connection Detailing Impacts Loads Transmitted to Shelter

Design of Storm Shelters | 65 | Shelters in Host Buildings

305.3 Other debris hazards. Lay down, rollover and collapse hazards shall be considered by the design professional when determining the location of shelters on the site.

▪ Commentary indicates that if the shelter can’t be sited to avoid these hazards, then impact loads should be considered, but no guidance provided – this problem gets fixed in ICC 500-2020, discussed later

Design of Storm Shelters | 67 | Windborne Debris - Tornado

305.1 Wind-borne debris. All shelters shall be designed for the impact of windborne debris in accordance with this section. 305.1.1 Missile criteria for tornado shelters. The debris impact test missile for all components of the shelter envelope of tornado shelters shall be a 15-pound (6.8 kg) sawn lumber 2 by 4 traveling at the speeds shown in Table 305.1.1.

Design of Storm Shelters | 69 | Windborne Debris - Hurricane

305.1.2 Missile criteria for hurricane shelters. The debris impact test missile for all components of the shelter envelope of hurricane shelters shall be a 9- pound (4.1 kg) sawn lumber 2 by 4. The speed of the test missile impacting vertical shelter surfaces shall be a minimum of 0.50 times the shelter design wind speed. The speed of the test missile impacting horizontal surfaces shall be 0.10 times the shelter design wind speed.

▪ 9 lb sawn lumber 2x4 ▪ Horizontal missile speed (impacting vertical surfaces) – 0.5 x V ▪ Vertical missile speed (impacting horizontal surfaces) – 0.1 x V

V = hurricane shelter design wind speed

Design of Storm Shelters | 70 | Component Design and Testing

▪ ICC 500 Section 306 ▪ Wall and roof assemblies must meet both pressure and missile impact requirements ▪ Openings must either: – Be debris impact resistant, or – Have impact resistant coverings (i.e., opening protectives) ▪ Chapter 8 provides test methods for impact and pressure testing

Design of Storm Shelters | 71 | FEMA P-361 THIRD EDITION / 2015

1-Day Training Introduction and A1 Module NOTE - The following several slides in impact testing and resistance are excepts from this FEMA safe room training class Wall and Roof Impact Tests

▪ No adequate method to model complexity of impacts ▪ Test specimen must be the exact design as the wall or roof that will be used on the safe room, including: ● Type, size, thickness of materials ● Type, size, and spacing of fasteners ● Configuration of all components ▪ For more details, see Wall Sections That Passed Previous Missile Impact Tests on FEMA’s safe room website

FEMA P-361 1-Day Training – Module B8 – Test Methods for Impact and Pressure Testing 73 Impact Resistance of Wood Frame Wall Assemblies

▪ The most effective wood frame designs incorporate masonry infill in the wall cavities or 14-gauge steel sheathing panels as the final layer in the assembly ▪ Which side of the wall the sheathing or protective material is attached to and the method of attachment can affect the performance of the wall in resisting wind-borne debris ▪ Improper configurations that can lead to perforation: ● 14-gauge or lighter steel backed by a substrate that prevents steel deflection ● 14-gauge or lighter steel sheets between plywood layers

FEMA P-361 1-Day Training – Module B8 – Test Methods for Impact and Pressure Testing 74 Wood Frame Wall Assemblies with Steel Sheathing: Examples That Passed Previous Missile Impact Tests

FEMA P-361 1-Day Training – Module B8 – Test Methods for Impact and Pressure Testing 75 Reinforced CMU Wall Assemblies: Examples That Passed Previous Missile Impact Tests

Note: More reinforcing steel may be required to resist wind loads

FEMA P-361 1-Day Training – Module B8 – Test Methods for Impact and Pressure Testing 76 Reinforced Concrete Wall Assemblies: Examples That Passed Previous Missile Impact Tests

Note: These wall assemblies may be impacted on either face.

FEMA P-361 1-Day Training – Module B8 – Test Methods for Impact and Pressure Testing 77 Wall Assemblies Missile Impact Test Video

FEMA P-361 1-Day Training – Module B8 – Test Methods for Impact and Pressure Testing 78 Performance of Door Assemblies

▪ Door assembly includes: Door, vision panel, hardware (locks and hinges), frame, and anchors into the safe room wall ▪ Installed door assembly must be the same type, size, configuration of materials, and swing direction used during testing ▪ Assemblies are debris-impact tested on the side of the door facing the exterior of the safe room ▪ Assemblies are pressure tested from both sides with positive pressure ▪ Residential Safe Room Door Fact Sheet: www.fema.gov/media- library/assets/documents/99139

FEMA P-361 1-Day Training – Module B8 – Test Methods for Impact and Pressure Testing 79 Performance of Glazed Opening Assemblies

▪ Glazed opening assembly includes: ● Glazing, glazing frame, anchors into the safe room wall ▪ Installed glazing assembly must be the same type, size, and configuration of materials used during testing ▪ Debris-resistant glazing is laminated glass, polycarbonate, or a combination of these materials ▪ Glazing is permitted to break during testing provided that: ● The missile does not perforate the glazing ● The glazing remains attached to the frame ● Ejected glass fragments do not perforate the witness screen

FEMA P-361 1-Day Training – Module B8 – Test Methods for Impact and Pressure Testing 80 Impact-Protective Systems

▪ Impact-protective systems include shutters, doors, shields, and cowlings ▪ Protect openings at louvers, grates, grilles, precast panel joints more than 3/8 inch wide, plumbing vents, roof drains, and emergency generator exhaust vents ▪ Non-operable, permanently affixed shields or cowlings do not need to be pressure tested, but are required to be tested for resistance to missile impact

FEMA P-361 1-Day Training – Module B8 – Test Methods for Impact and Pressure Testing 81 Presentation Outline

Design of Storm Shelters | 82 | Summary of IBC/IRC/IEBC Shelter Requirements

▪ 2009 IBC/IRC (and all later editions) – IF building a storm shelter, requirement to comply with ICC 500-2008 ▪ 2015 IBC (section 423) – Installation of tornado shelters required in the 250 mph tornado wind zone, for new • Schools • Fire, rescue, ambulance, and police stations • 911 call centers and emergency operations centers ▪ 2018 IBC/IEBC – Expanded requirements to explicitly include new buildings on existing school campuses, and additions to existing school buildings, over a certain size – Occupant capacity requirements for the shelter to house the full population of the school (with exceptions)

Design of Storm Shelters | 83 | 2018 IBC Shelter Requirements

Table 1604.5 Risk Category IV • Designated earthquake, hurricane or other emergency shelters

Design of Storm Shelters | 84 | 2018 IBC Shelter Requirements

Design of Storm Shelters | 85 | 2018 IBC Shelter Requirements

Maximum population likely to be in the school at any time No requirement to increase size of new building

Approximately 5 minutes travel time – consistent with FEMA guidance

Design of Storm Shelters | 86 | 2018 IEBC Shelter Requirements

▪ Added shelter requirements parallel to IBC Section 423.4 (for Group E Occupancies) into the International Existing Building Code – Applies to additions to existing buildings – Prevents gaming of the system to avoid shelters by adding to a building rather than constructing a new building on existing campus – 423.4 requirements apply to all 3 methods • Work Area Compliance Method-Additions: Section 1106

• Performance Compliance Methods 1301.2.3.1 Both sections • Prescriptive Compliance Method 502.8 point to 1106

Design of Storm Shelters | 87 | Direct Adoption of ICC 500 by States/Territories

▪ Alabama – Requirements for new K-12 schools to have ICC 500 tornado shelters – Requirements for certain higher education buildings to have tornado shelters ▪ Illinois – Requirements for new K-12 schools to have ICC 500 tornado shelters ▪ Puerto Rico – Requirements for new K-12 schools to have ICC 500 hurricane shelters

Design of Storm Shelters | 88 | Common Design Strategies for School Shelters

▪ Gymnasium ▪ Assembly Center ▪ Classroom Wing ▪ Hallways – retrofit possibilities for existing schools

Source: FEMA P-361 3rd ed.)

Design of Storm Shelters | 89 | Presentation Outline

Design of Storm Shelters | 90 | FEMA P-361 THIRD EDITION / 2015

1-Day Training Introduction and A1 Module

The following several slides are excepts from this FEMA safe room training class Codes/Standard vs. FEMA Building Science Publications

Building codes and standards ▪ Written by technical expert panels/committees ▪ Consensus documents, publically balloted ▪ When adopted, may be enforced by the building department ▪ Can only reference other consensus codes and standards ▪ Products may be approved through code “evaluation services” ▪ Laboratories and testing agencies provide product approvals and compliance certificates

FEMA P-361 1-Day Training - Introduction and Module A1 92 Codes/Standard vs. FEMA Building Science Publications (cont.)

Building Science Publications ▪ Written by technical expert panels/committees ▪ Evaluated by Review Committees, not publically balloted ▪ Typically, not adopted and therefore, not enforceable by the building department ▪ If adopted through special legislative rule or action, guidance may be enforced by the building department (rare) ▪ Only when part of a grant program does the design criteria change from “should” to “shall”

FEMA P-361 1-Day Training - Introduction and Module A1 93 Organization

Part A: Information that safe room planners, designers, owners, operators, and emergency management officials may find useful in planning, designing, and operating a safe room Part B: Arranged so the chapter order and content are in sequence with the first eight chapters of ICC 500

FEMA P-361 1-Day Training - Introduction and Module A1 94 Safe Room Tools and Resources: Website

▪ http://www.fema.gov/safe-room-resources ▪ Safe Room Door Assemblies and Components ▪ Example Operations and Maintenance Plans for Community Safe Rooms ▪ FEMA P-361 History and Relevant FEMA Building Science Activities ▪ Legacy FEMA P-361 Case Studies ▪ Previous Missile Impact Tests for Wood Sheathing ▪ Wall Sections That Passed Previous Missile Impact Tests

FEMA P-361 1-Day Training - Introduction and Module A1 95 Safe Room Tools and Resources: Website (cont.)

▪ http://www.fema.gov/safe-room-resources ▪ Safe room funding ▪ Publications and documents available for download include: ● Fact Sheets including Residential Tornado SR Door FS ● Flood Hazard ‘Quick Guides’ ● Post-disaster Reports and Recovery Advisories ▪ FEMA P-388: Safe Room Resources CD ▪ Much more!

FEMA P-361 1-Day Training - Introduction and Module A1 96 Safe Room or Storm Shelter?

▪ Both safe rooms and storm shelters provide life-safety protection from hurricanes and tornadoes ▪ Storm shelters must comply with ICC 500 ▪ Safe rooms must comply with ICC 500 and FEMA P-361 which is higher for some criteria ▪ Because safe rooms comply with ICC 500, they also qualify as storm shelters

FEMA P-361 1-Day Training - Introduction and Module A1 97 What Is a Safe Room? (cont.)

▪ Meets or exceeds ICC 500 requirements and FEMA’s Recommended Criteria described in FEMA P-361 Part B ▪ Because safe rooms comply with ICC 500, they also qualify as storm shelters

FEMA P-361 1-Day Training - Introduction and Module A1 98 Background on FEMA Safe Rooms

▪ Post-disaster studies have been conducted since the 1970s to determine safe room design ▪ FEMA technical building science teams observe and assess building performance after disasters of national significance in the United States ▪ FEMA P-320 was first released in 1998 based on team observations and research from Texas Tech University ▪ The first edition of FEMA P-361 was released in 2000 ▪ FEMA was involved with the development of ICC 500, the first consensus code for storm shelters released in 2008

FEMA P-361 1-Day Training - Introduction and Module A1 99 FEMA P-361 and ICC 500 Comparison

▪ Safe rooms constructed with FEMA grant funds are required to adhere to FEMA Recommended Criteria described at the beginning of Part B chapters as well as the corresponding ICC 500 requirements ▪ FEMA P-361 provides recommended criteria and best practices, while ICC 500 is a minimum standard

FEMA P-361 1-Day Training - Introduction and Module A1 100 FEMA P-361 and ICC 500 Comparison (cont.)

▪ FEMA P-361 addresses emergency management considerations; ICC 500 does not ▪ FEMA P-361 includes guidance for emerging issues and concerns learned through FEMA technical building science team observations, assessments, and recommendations

FEMA P-361 1-Day Training - Introduction and Module A1 101 Differences Between P-361 and ICC 500

FEMA P-361 3rd ed. APPENDIX D

Additional differences (not shown here) • Flood-related criteria o Minimum floor elevations o Siting

• First aid kit requirements

Safe Rooms | 102 | Types of Safe Rooms

Residential safe room: Serves occupants of dwelling units and has an occupant load not exceeding 16 persons Community safe room: Any safe room not defined as a residential safe room

FEMA P-361 1-Day Training - Introduction and Module A1 103 Types of Safe Rooms (cont.)

Stand-alone safe room: A separate building designed and constructed or retrofitted to withstand extreme winds and the impact of wind-borne debris during tornadoes, hurricanes, or other extreme-wind events. Internal safe room: A specially designed and constructed room or area within or attached to a larger building. It should be structurally independent of the larger building, but provide the same wind and wind-borne debris protection as a stand-alone safe room.

FEMA P-361 1-Day Training - Introduction and Module A1 104 Types of Safe Rooms (cont.)

Above-ground safe room Below-ground safe room

FEMA P-361 1-Day Training - Introduction and Module A1 105 Types of Safe Rooms (cont.)

Prefabricated safe room: A safe room that has been assembled off-site and transported to the site where it will be installed Site-built safe room: Any safe room not defined as prefabricated safe room

FEMA P-361 1-Day Training - Introduction and Module A1 106 Deciding Whether to Install a Safe Room

Considerations include: ▪ The likelihood of an area being threatened by an extreme-wind event ▪ The vulnerability of a structure to an extreme-wind event ▪ The risk or potential losses (deaths and injuries) associated with an extreme-wind event ▪ The cost of constructing a safe room

FEMA P-361 1-Day Training - Introduction and Module A1 107 Deciding Whether to Install a Safe Room (cont.)

FEMA P-361 1-Day Training - Introduction and Module A1 108 Benefit-Cost Analysis

▪ BCA is used to estimate the cost-effectiveness of a proposed project ▪ FEMA HMA requires a Benefit-Cost Ratio of 1.0 or greater for a project to be eligible for funding

FEMA P-361 1-Day Training – Module A3 – Costs and Benefit-Cost Analysis 109 Benefit-Cost Analysis Software

▪ Focused exclusively on reduction of injuries and deaths ▪ Modules: ● Tornado Safe Room BCA ● Hurricane Safe Room BCA ▪ Modules developed for community and large residential tornado and hurricane safe rooms ▪ www.fema.gov/benefit-cost-analysis

FEMA P-361 1-Day Training – Module A3 – Costs and Benefit-Cost Analysis 110 Presentation Outline

Design of Storm Shelters | 111 | FEMA / FIMA Hazard Mitigation Assistance Grants

Hazard Mitigation Assistance Grants

| 112 | FEMA Funding for Safe Rooms As of 11/2020: $1.24B - 40,554 Residential / 2,173 Community ▪ $1.09B Community and $147.3M Residential

As of 01/2015: $984M - 25K+ Residential and 2,000 Community Safe Rooms

As of 2010: $460M - 20K+ Residential and 1,000+ Community Safe Rooms ▪ Total funds obligated for community safe rooms – Over $405M (Federal share) – 1,000+ community safe room projects ▪ Total funds obligated for residential safe rooms – Over $55M (Federal share) – 20,000+ residential and in-residence safe rooms

| 113 | Eligible Activity: FEMA 361 Safe Rooms - Community and Residential Hurricane and / or Tornado

| 114 | Safe Room Eligibility Criteria

▪ PDM and HMGP funds may only be used for safe room projects designed to achieve “near-absolute protection” as described in the current editions of FEMA P-320, Taking Shelter From the Storm: Building a Safe Room For Your Home or Small Business (2014), and FEMA P-361, Safe Rooms for Tornadoes and Hurricanes: Guidance for Community and Residential Safe Rooms (2015). ▪ PDM and HMGP funds are not available for general population shelters, including evacuation and recovery shelters. ▪ Documentation that demonstrates compliance with local planning, zoning, building, and other applicable codes ▪ Identification of the impacted population – the population to be protected ▪ Travel limitations – Tornado – 5 min. / 0.5 mi. ▪ A cost-effectiveness analysis using approved FEMA methodology (BCA) – Pre-Calculated Benefits for Residential Safe Rooms ▪ Description of the approach the subapplicant will use in preparing the O&M plan

| 115 | Useful Links and Resources FEMA Safe Room Helpline: [email protected] FEMA Safe Room Hotline: 1-866-927-2104 State Hazard Mitigation Officers: http://www.fema.gov/state-hazard-mitigation-officers

FEMA’s Hazard Mitigation Assistance (HMA) Further info on HMA funding for Safe Rooms, visit: https://www.fema.gov/hazard-mitigation-assistance

| 116 | Presentation Outline

Design of Storm Shelters: ICC 500-2020 | 117 | New Edition: ICC 500 - 2020

▪ Many Significant Changes – Committee considered over 150 proposals, resulting in modifications to • Scope • Listing and labeling • Peer review • Inspections • Updated to ASCE 7-16 as basis for load provisions • Load combinations • Hurricane shelter design wind speed map • Laydown and falling debris hazards • Flood elevation requirements • Fire protection • Impact and pressure testing • New Appendix - Shelter Emergency Ops Plan • Many others ▪ Schedule – Published December 2020 (version w/ commentary coming early 2021) – Incorporation by reference into the 2021 I-Codes ▪ Revised editions of FEMA Safe Room pubs to follow shortly Design of Storm Shelters: ICC 500-2020 | 118 | Load Combinations

▪ ICC 500 Loads and Load Combinations are in addition to loads and load combinations required in the applicable building code

Strength Design 2018 IBC (excluding F, H, and E) ICC 500-2020 Tornado Shelter

1.4D 1.4D

1.2D + 1.6L + 0.5(Lr or S or R) 1.2D + 1.6LT + 0.5LrT

1.2D + 1.6(Lr or S or R) + (f1L or 0.5W) 1.2D + 1.6LrT + (LT or 0.5WT)

1.2D + 1.0W + f1L + 0.5 (Lr or S or R) 1.2D + 1.0WT + LT + 0.5LrT

0.9D + 1.0W 0.9D + 1.0WT Modifications • Tornado specific live, roof live, wind loads Similar changes for ASD Combinations • No rain or snow load included in combinations

Design of Storm Shelters: ICC 500-2020 | 119 | Laydown and Falling Debris

▪ Significant changes from 2014 edition – Revised terminology and definitions – Requirements explicit treatment and determination of impact loads, not just ‘siting’ considerations

Falling Debris Hazard Laydown Hazard

Source: FEMA

Design of Storm Shelters: ICC 500-2020 | 120 | Laydown Hazard Definition

▪ Laydown Hazard. Adjacent building elements, other structures and natural objects, that could fall onto the roof of a storm shelter, such as exterior walls of adjacent single story structures, self-supporting towers, poles or large trees.

Source: Marc Levitan/NIST Source: Barbara Watson, NOAA/NWS

Design of Storm Shelters: ICC 500-2020 | 121 | Falling Debris Hazard Definition

▪ Falling Debris Hazard. Exterior components, cladding, and appurtenances, such as parapet walls, masonry cladding, or rooftop equipment, that could fall onto the roof of a storm shelter from wind damage to adjacent, taller buildings or taller sections of a host building.

Source: FEMA Source: FEMA

Design of Storm Shelters: ICC 500-2020 | 122 | Laydown Radius

▪ Shelter roofs must be designed for impact loads (with minimum impact factor of 2.0) – If within laydown radius for laydown hazard or fall radius of falling debris hazard

Falling Debris Radius

Source: FEMA Source: FEMA

Design of Storm Shelters: ICC 500-2020 | 123 | Presentation Outline

Design of Storm Shelters | 124 | Bonus: ASCE 7-22 Tornado Loads Preview

Tornado Load Chapter coming to ASCE 7-22 – Developed by NIST, Applied Research Associates, and the ASCE 7 Tornado Task Committee – New tornado hazard maps, at return periods consistent with ASCE 7 reliability targets – Vertical profile of horizontal component of tornado winds – Tornado pressure and load coefficients • Effects of vertical component of wind handed through a modification factor on roof pressure coefficients – Procedures to determine tornado loads on • Main Wind Force Resisting System (MWFRS) • Components and Cladding (C&C)

ASCE 7-22 Tornado Loads | 125 | Example Tornado Speed (VT) Maps

Risk Category III Risk Category IV (1,700 Year) (3,000 Year)

84 99 Maps for 8 plan area sizes from 1 to 4M sq ft

111 125 Tornado loads can begin to control over wind loads

when VT > 0.5V Tornado speeds are 3-s peak gusts in mph at 10 m height

ASCE 7-22 Tornado Loads | 126 | Tornado Load Methodology

• Built on ASCE 7-16 Wind Load Framework • Worked closely with mobile radar community to analyze radar-measured tornado wind speeds and develop tornado velocity pressure profile, consistent with assumptions used in development of tornado hazard maps • Developed New and Modified Wind Load Factors for Tornadoes

o Tornado Directionality Factor KdT, for MWFRS and C&C

o Internal Pressure Coefficient GCpiT, including effects of Atmospheric Pressure Change o Tornado Pressure Coefficient Adjustment Factor

for Vertical Winds KvT, for MWFRS and C&C

ASCE 7-22 Tornado Loads | 127 | Tornado Provisions

▪ New Chapter 32 – Proposed new chapter on Tornado Loads – Includes tornado maps at return periods consistent with reliability requirements for Risk Category III and IV buildings and structures (including places of public assembly, and critical and essential facilities) ▪ New Chapter 32 Appendix – Proposed new Appendix with long return period Tornado Maps to support the nuclear industry and performance-based design – 10,000 to 10,000,000 year Mean Recurrence Intervals, for point target through 4M sq ft target size

ASCE 7-22 Tornado Loads | 128 | Opportunities to Stay Informed/Participate

ASCE 7-22 Status NOAA Photo Library - OAR/ERL/National Severe Storms Laboratory (NSSL) • Ballots for the new Tornado chapter and for associated changes in other chapters passed the ASCE 7 Main Committee • Currently working through resolution of comments

All ASCE 7 Committee meetings are public • meeting schedule: https://www.asce.org/structural-engineering/asce-7- and-sei-standards/ Public Comment Draft 7-22 • June-July of 2021 • 45-day comment period • All comments must be considered and responded to by the Main Committee

Implications for ICC 500-2023

ASCE 7-22 Tornado Loads | 129 | Presentation Outline

Design of Storm Shelters | 130 | Additional Resources – Shelters and Safe Rooms

▪ ICC 500 Standard + Commentary – 2014 ed.; (2020 ed. available spring 2021) ▪ FEMA P-320 (free download) – 4th ed.; (5th ed available soon) ▪ FEMA P-361 (free download) – 3rd ed.; (4th ed. available soon) ▪ FEMA Safe Room Hotline: – 1-866-927-2104 ▪ FEMA Safe Room Help Line – [email protected] ▪ FEMA Grants – www.fema.gov/grants/mitigation – 1-866/222-3580

Design of Storm Shelters | 131 | Additional Resources – Vulnerability Assessment

▪ Guidelines for vulnerability assessment of critical facilities affected by hurricanes, tornadoes, and other windstorms ▪ Methods to assess vulnerability to – wind pressure – wind-borne debris – wind-driven rain

Design of Storm Shelters | 132 | Additional Resources – Tornado Effects/Basis for Shelters

The first tornado study to include storm characteristics, building performance, emergency communication and human behavior together - with assessment of the impact of each on fatalities.

Focused reconnaissance on 2 schools and 1 hospital

http://dx.doi.org/10.6028/NIST.NCSTAR.3

https://doi.org/10.6028/NIST.sp.1164

Design of Storm Shelters | 133 | FEMA Mitigation Assessment Teams

1999 Oklahoma and 2011 Alabama and 2014 Moore, Oklahoma Kansas (FEMA 342) Missouri (FEMA P-908) (FEMA P-1020)

FEMA P-361 1-Day Training - Introduction and Module A1 134 Closing Thoughts – Code and Standard Requirements

ICC 500 Purpose: The purpose of this standard is to establish minimum requirements to safeguard the public health, safety and general welfare relative to the design, construction and installation of storm shelters constructed for protection from tornadoes, hurricanes and other severe windstorms.

ICC 500-2020

Design of Storm Shelters | 135 | Closing Thoughts – Moving Beyond Code Minimums

St. Johns Regional Medical Center, Joplin MO, May 22, 2011

Source: NIST

Source: NIST Consideration of other building performance objectives beyond life safety • Buildings intended to remain operational in the event of extreme environmental loading from flood, wind, snow, or earthquakes (ASCE 7 Essential Facilities) • Immediate Occupancy © 2011 Malcolm Carter. Used with Permission.

Design of Storm Shelters | 136 | Storm Shelter Design: Overview of IBC and ICC-500 Requirements Including Significant Changes in ICC 500-2020

University of Minnesota Structural Engineering Webinar 1/26/2021

Questions / Comments?

Marc L. Levitan [email protected]