8.2.3 Characterizing Motor Vehicles.
Motor vehicles (cars, buses, trucks, etc.) have fairly consistent parameters, except for size. They usually have flat sheet metal roofs that are approximately 0.04-0.05 inches thick. The sides have inner and outer metal panels. The front windshield is made of laminated glass, while the back and side windows are usually tempered glass. The vehicle speed should also be accounted for in both the impact kinetic energy, impact angle, and the projected area (an analog to “running in the rain”). For debris analyses, inert debris can puncture the roof or break the windows. Explosive debris can break windows, or in severe cases, cause the vehicle to become overturned. For toxic analyses, the analyst should assume this structure is leaky. For blast analyses, the likely failure modes are window breakage and collapse of the roof. In severe blast cases, the vehicle can become overturned.
8.2.4 Characterizing Waterborne Vessels.
When considering effects of sheltering for waterborne vessels, applicants should account for the differences between ground vehicles and waterborne vessels. If a waterborne vessel is hit and sinks, all people onboard the vessel are in danger of a casualty or fatality, no matter if they are sheltering or not. Applicants should also consider the variety of waterborne vessels that may be present in the area during launch (from cruise ships to fishing boats and varieties in between). This will determine the structure and material of the waterborne vessel. Small fishing boats can be made of wood, while larger fishing boats can be made of steel. Larger waterborne vessels are usually made of steel. Inert debris can hit people standing on deck of the waterborne vessel. The RCC 321-20 Supplement4 provides acceptable vulnerability models for people in various types of waterborne vessels.
8.2.5 Characterizing Other Vehicles.
In certain situations, people in other vehicles, such as trains, can be at risk. These should be characterized adequately for risk modeling.
8.2.6 Characterizing Structures for Building-by-Building Modeling.
For on-facility populations, the resolution for structures should be defined per single entity or building. However, there are some cases where the population model should divide a building into sections. These cases include:
1. When the construction characteristics change significantly from one section of a building to another (e.g. part of the building has reinforced block walls while another part is a light metal construction);
2. For different floors of a building (typically for multi-floor buildings, only the first few floors from the top of the building experience significant risk; all other floors below can be categorized the same);
4 Risk Committee, Range Safety Group, Range Commanders Council, Common Risk Criteria for National Test Ranges, RCC 321-10, White Sands, NM 2010.
3. When window types, sizes, and amounts vary significantly from one section of the building to another; or
4. When the geometry changes significantly from one section of the building to another (e.g. an L-shaped building).
Note: The applicant should also consider when using the collected data to allocate some percentage of the people to be outside the building or in vehicles next to the building.
Usually this is between 1 and 10 percent, depending on the building usage and time of day, but people may go outside to watch a mission.
8.2.7 Characterizing Sheltering in Near-facility Areas.
For near-facility populations, the resolution for structures should be defined by groups of buildings. Groupings include housing tracts, business parks, shopping areas (can include gas stations, restaurants, stores), and recreation areas. Within these groups, the applicant should designate what percentage is a certain building type (e.g. what percentage are single family homes. Structures with high population density, such as hospitals, schools and hotels, should be specifically categorized similar to how on-facility buildings are.
8.2.8 Characterizing Sheltering in Larger Regions.
For larger regions, demographic data should be used to determine sheltering categories.
Examples on how to generalize building structures by these categories with data available can be found in paragraph 8.2.6 of this AC. AIAA Report #2005-63225 provides a discussion of how to use demographic data to estimate sheltering. The following is an extremely simplified example of how to calculate sheltering percentages. One population center has a population where all people are working.
There are two occupation categories: office workers (assume 60 percent) and agricultural workers (assume 40 percent). Assume the office workers are always inside.
The buildings they occupy could either be light (e.g. wood), medium (e.g. steel), or heavy (e.g. concrete) sheltering, with an equal probability assumption (about 33 percent each sheltering type). Note that a typical analysis would normally have many more structure categories. Assume the agricultural workers are only inside 25 percent of the time, and when they are inside, they are in light structures. The sheltering distribution can be calculated by the following equation:
5 Larson, Erik, Large Region Population Sheltering Models for Space Debris Risk Analysis, American Institute of Aeronautics and Astronautics (AIAA) Atmospheric Flight Mechanics Conference and Exhibit, Report #2005 -6322, San Francisco, CA, August 2005.
Office Agricultural cwood cconrete � 100% Wood 33% 25% 60% office � � = � �Concrete 33% 0%� � �csteel at work � 40% agriculturalSteel 33% 0%copen Open 0% 75% 10/12/2022 AC 450.123-1 This equation results in:
𝑐𝑐𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤 = 100% × (33% × 60% × 25% × 40%) = 30% of the population in wood structures, 𝑐𝑐𝑐𝑐𝑤𝑤𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 = 100% × (33% × 60% × 0% × 40%) = 20% of the population in concrete structures, 𝑐𝑐𝑠𝑠𝑐𝑐𝑐𝑐𝑐𝑐𝑠𝑠 = 100% × (33% × 60% × 0% × 40%) = 20% of the population in steel structures, 𝑐𝑐𝑤𝑤𝑜𝑜𝑐𝑐𝑐𝑐 = 100% × (33% × 60% × 75% × 40%) = 30% of the population in the open.
Actual calculations by the applicant will be much more complex, with more population and structure categories.
8.3 Input Data Sources.
Section 450.123(b)(3) requires applicants to use reliable, accurate, and timely source data. Reliable source data means good quality data that can be trusted. In some parts of the world, some of what is considered “official” data is inaccurate due to either poor data collection or intentional deception. Accurate source data means data that is of sufficient or appropriate resolution. Sufficient resolution is discussed in paragraph 8.1.1 of this AC. Timely source data means data that is relatively up-to-date. Data sources are only updated periodically. For example, the U.S. census is only updated every 10 years.
It is important that the applicant adjusts the data they are using for the correct time period. It is possible that no single data source will have all of these qualities.
Applicants may need to use multiple input data sources or make adjustments to the data to satisfy the three requirements in § 450.123(b)(3).
8.3.1 Site Operator Data.
Operators should not presume that a launch or reentry site or Federal site will provide population data, either for the facility or for the near area. If an operator desires this data, that may be accomplished via an agreement between the operator and the site.
An agreement with a Federal site might not provide population data to the launch operator, but rather the Federal site may conduct the analysis as part of range safety services provided to the launch operator. Manual surveys of individual buildings may be performed to collect population and structure information for facilities on site or within a few km of the site boundary.
8.3.2 Completeness of Databases.
To comply with § 450.123(b)(3), several products (databases and software applications) are publicly available to assist in the development of a population model for risk analysis. Note that these products may be insufficient for identifying populations at specific locations (e.g. restaurants, viewing sites, etc.), particularly on launch or reentry day/night/weekend/time-of-day/etc., and may not communicate proprietary or security information (e.g. the number of people in a building at a military facility, in a company’s processing facility during a work shift, etc.). These populations and buildings may be some of the closest to the hazard, thus critical to the risk assessment, and may need to be individually surveyed to provide the level of fidelity required to meet § 450.115(b).
8.3.3 On-Facility Data.
On-facility population data resolution should be very high, and thus this data is usually obtained by surveying the area, if safety and security allow. When surveying on-facility populations, there should be two main focuses: building parameters, and number of people vs. time and classification (per paragraph 8.3.3.2 of this AC). The data should be validated to ensure every building has been captured. This may be accomplished by overlaying the buildings on satellite imagery, such as in Google Earth. There may be cases where an on-facility survey is not possible (e.g., for security reasons). In cases like this, the facility operator should perform the survey and provide the data to the applicant.
8.3.3.1 Building parameters.
Applicants should record the construction characteristics (e.g. frame, roof, windows, etc.), geographical coordinates, and area of the building. It may be helpful during the survey to draw or take pictures of the buildings.
Building data should be recorded by floor, if possible. Examples of surveys done for a building, a section of a building, and a floor within the building can be seen in Figure 4, Figure 5, and Figure 6 of this AC. This example shows a survey of an on-base fire station. The fire station, while all one building, is broken down into different sections. Figure 4 recognizes that the building is “Very Loose” in terms of air exchange rate, indicating that a toxic exposure would infiltrate the building quickly.
Figure 5 focuses on a single section of the building. It notes what the section of the building is used for, which helps with estimates of how many people will be in this section of the building during certain time periods. The surveyor also noted which side of the section that windows were on. A high-fidelity population model should go a step further and record the number and dimensions of the windows. Figure 5 also notes how many floors the section has. Figure 6 records the wall type, roof type, and frame type of the specific building section.
Building Data Sheet Sheet ___!_ of -- Recorded By: )de Building ID: --- Date: 9/3/05 latitude (center):--l ongitude (center):- Photos: JL Yes - No Building Description: F119 Station Base Area: HQ General Usage : Storage of high profile whlclN, edmlnlatmlon, stonige No. Of Sections :- .;:4;..._ __ _ Exchange Rate ---- #~ ho.Jr or, Year Built: ??? D Air Tight O l oose Iii Very l oose Specialty Bldg: 0 Trailer/Mod O Blockhouse O Vert Assy O Oil Rig O Auto Ske cl :
•w -· ~ u .... h 4 "'C> lh•I ...
1::1..1 ..
~ - . - ' --,
r.:. 1 - ., ..
.....
-- I I 10/12/2022 AC 450.123-1 Figure 4. Survey Example for Building Data Sheet 25
Section Data Sheet Sheet ___!_ of __ Building ID: ~ Section No.: __ 1 __ _ Recorded By: Jdc Date: 913/0!l Latitude (center): ?'I? Longitude (center): ?'I?
Section Description: Donnltory General Usage: Uvlng Space
Length (feet): __ aa__ Width (feet): _ ....;;;28;;_
Azimuth (length rel. to North) Number of Floors: _ ..;1 __ __ eo ___ degrees Total Height: __ 9 __ feet Sketch :
I :111111 3 11W n I Ill • rH 0,1 I I & .. • - ~ ~ ~·. - -.. .. , II .. , V _,_ I I "· - > , 10/12/2022 AC 450.123-1 Figure 5. Survey Example for Section Data Sheet 26
n-Base Floor Data Sheet Sheet __ 3_ of __ Recorded By: ldc Building ID: - SectNo .: _ 1_ Floor No.: 1 Date: 913/0S Description: Dormitory Organization:
Height: _....;II __ feet Wall Type 1 )--Wood Stud
2)--Metal Stud
3) __ Unrein! Brick
4)~ Unrein! Block
5) __ Reinf Block
6)--Light Metal E1°€ 1Jyp0
1)~ Builtup/Composite
2)-- Light Wood
3)--Wood Panelized
7)--Heavy Metal
8)--Reinforced Concrete
9)--RC Tiltup 4 l -- Lt-Wt Metal Deck on SU Joist
5)--Lt-Wt Cone on Metal Deck
6)- Heavy Metal on SU Girder
7)--Reinf Cone on Metal Deck
8)-- Reinforced Concrete
9)-- Tile
10)--Unknown 10)-- Unknown Frame Type 1 )--Wood Stud (bearing wall)
2)--Metal Stud (bearing wal )
3)--Steel Braced
4)--Steel Moment Resisting
5) ~ Concrete Moment Resisting
6)- Unknown 10/12/2022 AC 450.123-1 Figure 6. Survey Example for Floor Data Sheet 27
8.3.3.2 Number of People.
The second focus or survey is number of people. This may be done by a physical counting of occupants in the building at different times of the day. If this is not possible, the applicant should obtain the information from the facility operator. It is important, during the survey, to also get an idea of the variability of the on-facility population with time. The survey should specify when people are in the building during daytime, nighttime, weekdays, or weekends. More temporal variation data to consider recording during the survey is discussed in paragraph 8.1.2 of this AC.
The people also should be categorized as personnel involved in supporting a launch or reentry, NOP, or other public, and whether they are government personnel (for MPL considerations). When collecting data for NOP, applicants either should be conservative or work with site operators to get up-to-date data on what NOP will be in the area during the operation. More specifically, the launch or reentry operator should work with the site operator (i.e., an operator of a Federal site or FAA-licensed launch or reentry site) and the operators of any neighboring sites to identify NOP personnel because the site operators are in the best position to identify which personnel are required to perform safety, security, or critical tasks at the launch site.
8.3.4 Spectator Areas.
To account for spectators, the applicant should determine where the designated spectator areas will be. The applicant may set their own spectator areas by working with the city or may work with the launch site to establish where spectators historically gather such as beaches, piers, or roadblocks. Due to the uncertainty of how many spectators could be in the area at the time, the applicant should consider calculating EC for different varying numbers of people in the area. Table 4 shows an example of the different population sizes.
Table 4. Example of EC for Varying Population Sizes in a Viewing Area People in Viewing Area Collective Risk (EC) per million 10,000 3 30,000 7 100,000 22 28
8.3.5 Near-Facility Population Data.
Near-facility population data does not have the same high resolution as on-facility population data. The methods to collect this data may include a mix of surveying and database use, depending on the distance from the launch or landing site (see paragraph 8.1.1.3 of this AC for more information on resolutions for near-facility populations). At near distances all individual buildings and locations where people gather should be identified. At mid-range distances, the applicant should identify critical population centers, such as schools and hospitals, as individual buildings, but lower density population areas may represent groups of buildings (e.g. 1 km2). At further distances, the size of lower density regions may be larger and clusters of buildings with higher density population may be used (e.g. a whole school campus).
8.3.5.1 General Data Sources.
Data may be collected by identifying structures using satellite imagery and tax assessor data. For specific buildings such as schools and hospitals, the applicant may need to survey the specific structures and contact building operators for occupancy data to provide the level of fidelity required to meet § 450.115(b). Sources for demographic data for the more generalized populations can be found in paragraph 8.3.5.2 of this AC. The LandScan USA product is very helpful for obtaining high-resolution day/night population data averages, although higher-fidelity temporal variation is not included.
8.3.5.2 Seasonal Variations.
Seasonal variations can be important factors that either alter distribution of people indoors or outdoors (e.g. more people outdoors in summer than winter) or alter total local population count (e.g. Chincoteague Island near the National Aeronautics and Space Administration (NASA) Wallops Flight Facility range has about 3,000 residents during off season winter months and 25,000 to 40,000 total population during the June-August tourist season with many of these people outdoors during the day and in local hotels at night).
5 - 1-95 (Brevard County) - TTMS # 700322 1600.00 1400.00 5 noo.oo £ 100000 K 80000 ~ 600.00 ~ 400.00
200.00 000 0322 - March 26, 2020 l 2 3 4 S 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Time of Day ■ Pre North ■ Pre South ■ Post North ■ Pos:t South 28 % Change N 36.3% S 37.0 % Foo'fi\ 'Lil ~ \) -CIVL IHTl!GRATEO MAHA04!MUff 10/12/2022 AC 450.123-1
8.3.5.3 Transportation Data.
Transportation modes should receive special consideration. Most public roads are not a significant contributor risk as compared to background populations, except in extremely sparely populated areas. A major road, such as a freeway, should be accounted for. Road usage information can be acquired from state Departments of Transportation and from the U.S. Bureau of Transportation.6 Most stated transportation departments will provide the volume of cars on a road for a particular day or time period (an example of this is shown in Figure 7 of this AC). Maritime traffic information can be acquired from the Department of Transportation SeaVision.7 Figure 7. Example Vehicle Volume per Hour data for Brevard County for March 26, 20208
8.3.6 Offshore Oil Rigs Population.
For certain sites, applicants should account for offshore oil rig populations. The population data for these sites may or may not be in a population database. Surveying these areas for population data may not be feasible. Instead, online gas and oil platform maps may be used to locate current locations around the launch site. A variety of 6 Bureau of Transportation Statistics, “Number of U.S. Aircraft, Vehicles, Vessels, and Other Conveyances”, https://www.bts.gov/content/number-us-aircraft-vehicles-vessels-and-other-conveyances, 2021.
7 https://info.seavision.volpe.dot.gov/ 8 Larson, Erik, Large Region Population Sheltering Models for Space Debris Risk Analysis, American Institute of Aeronautics and Astronautics (AIAA) Atmospheric Flight Mechanics Conference and Exhibit, Report #2005-6322, San Francisco, CA, August 2005.
open-source trackers exist for location of these structures; however, official data for U.S. launches are typically provided by the Bureau of Ocean Energy Management9 and the National Oceanic and Atmospheric Administration.10 Applicants should conduct a survey of these resources and contact the management of potentially exposed structures for up-to-date population data.
8.3.7 Large-Region Population Data.
For regions further from the operation launch or landing site, the population data do not need as high resolution as on-facility or near-facility data. Existing datasets are typically used to develop this data, as discussed in AIAA Report #2005-6322.4
8.3.7.1 Use of Demographic Data.
Useful products should be divided into two groups: (1) datasets consisting of spatial data to be integrated into a model, and (2) processing and display applications used to integrate and analyze datasets. Regarding the first category, overhead imagery, state, and/or local development records, including planning and building department and U.S. Census data,11 are often useful sources of population modeling data, including sheltering data. U.S. Census data specifies where people live and may be a source of demographic data, used to infer the distribution of people among various types of buildings. Local development records, such as a county tax assessor, may provide useful information on types of commercial buildings and zoning classification for different areas. Operators may use demographic data to assign population distributions to different classes of population groups (e.g., farmers, white-collar workers, and children) and as a function of time-of-day (workday/evening/weekend).
8.3.7.2 Sources of Demographic Data.
Table 5 of this AC provides a summary of three sources of data, which are accessible and commonly used for population modeling for flight safety analysis. All the sources in Table 5 are considered reliable, accurate, and timely sources of data consistent with § 450.123(b)(3). Information is available on potential sources of the baseline data necessary to construct a population model. Landscan12 and Gridded Population of the World (GPW)13 may be used to provide population estimates for grid cell areas as small as 1 square km covering the populated areas of the globe.
9 https://www.boem.gov/oil-gas-energy/mapping-and-data 10 https://www.ncei.noaa.gov/ 11 United States Census Bureau, “Explore Census Data”, https://data.census.gov/cedsci/, 2020.
12 Oak Ridge National Laboratory, “LandScan, Geographic Information Science and Technology”, https://landscan.ornl.gov/, 2018.
13 Socioeconomic Data and Applications Center (SEDAC), “Gridded Population of the World (GPW), v4, CIESIN, Columbia University, New York, New York, https://sedac.ciesin.columbia.edu/, 2020.
Table 5. Comparison of Data Sources for Quantifying Populations at Risk Source of Data: LandScan GPW U.S. Census Accuracy Very good Good Excellent Resolution Very good Very good Excellent Availability Good Excellent Excellent Completeness Global Global U.S. Only Sheltering Information None None Very Good
8.3.7.3 Sheltering.
The sources above will not provide direct sheltering data for populations.
However, they will provide data that is applicable to sheltering.
To provide data that is applicable to sheltering, the applicant should use three categories to generalize building structures: home, work, and school.
8.3.7.4 United States (U.S) Census Data.
The U.S. Census14 provides data that alludes to materials and protection level of the home structure, such as number of units in a structure (1, 2 to 4, 5 or more, mobile, other). A 1-unit structure corresponds to a single-family residence, which typically in the U.S. have wood or tile roofs, while “5 or more” can correspond to apartment complexes, which typically have steel or concrete roofs. While the information is not available on a tract-by-tract basis, it is available for a larger general area.
Applicants may apply the percentages of the larger general area to the specific tracts of interest.
14 United States Census Bureau, “Explore Census Data”, https://data.census.gov/cedsci/, 2020.