Rosenheim, Nathanael, Roberto Guidotti, Paolo Gardoni & Walter Gillis Peacock. (2021). Integration of detailed household and housing unit characteristic data with critical infrastructure for post-hazard resilience modeling. Sustainable and Resilient Infrastructure. 6(6), 385-401. https://doi.org/10.1080/23789689.2019.1681821

Rosenheim, Nathanael. (2022). "Detailed Household and Housing Unit Characteristics: Data and Replication Code." [Version 2] DesignSafe-CI. https://doi.org/10.17603/ds2-jwf6-s535

People are the most important part of community resilience planning. However, models for community resilience planning tend to focus on buildings and infrastructure. This repository provides a solution that connects people to buildings for community resilience models. The housing unit inventory method transforms aggregated population data into disaggregated housing unit data that includes occupied and vacant housing unit characteristics. Detailed household characteristics include size, race, ethnicity, income, group quarters type, vacancy type and census block. Applications use the housing unit allocation method to assign the housing unit inventory to structures within each census block through a reproducible and randomized process. The benefits of the housing unit inventory include community resilience statistics that intersect detailed population characteristics with hazard impacts on infrastructure; uncertainty propagation; and a means to identify gaps in infrastructure data such as limited building data. This repository includes all of the python code files. Python is an open source programming language and the code files provide future users with the tools to generate a 2010 housing unit inventory for any county in the United States. Applications of the method are reproducible in IN-CORE (Interdependent Networked Community Resilience Modeling Environment).

Center for Risk-Based Community Resilience Planning : NIST-70NANB20H008, NIST-70NANB15H044 (http://resilience.colostate.edu/)

Interdependent Networked Community Resilience Modeling Environment (IN-CORE) : https://incore.ncsa.illinois.edu/

Hazard Reduction and Recovery Center (HRRC): https://www.arch.tamu.edu/impact/centers-institutes-outreach/hazard-reduction-recovery-center/

Keywords: Social and Economic Population Data, Housing Unit, US Census, Synthetic Population

Output Data License: Open Data Commons Attribution License (https://opendatacommons.org/licenses/by/summary/)

Program Code: Mozilla Public License Version 2.0 (https://www.mozilla.org/en-US/MPL/2.0/)

Project-2961 - Available at: https://www.designsafe-ci.org/data/browser/public/designsafe.storage.published/PRJ-2961

• acg = api.census.gov - primary data source
• HRRC = Hazard Reduction and Recovery Center
• hui = housing unit inventory
• ncoda = Intersect Community Data
• IN-CORE = Interdependent Networked Community Resilience Modeling Environment

This product uses the Census Bureau Data API but is not endorsed or certified by the Census Bureau. https://www.census.gov/data/developers/about/terms-of-service.html

"Users will not use these data, alone or in combination with any other Census or non-Census data, to:

1. identify any individual person, household, business or other entity
2. not link or combine these data with information in any other Census or non-Census dataset in a manner that identifies an individual person, household, business or other entity
3. not publish information from these data files, particularly in combination with any other Census or non-Census data, in a manner that identifies any individual person, household, business or other entity; and
4. not use the identity of any person or establishment discovered inadvertently and advise the Director of the Census Bureau of any such discovery by calling the U.S. Census Bureau’s Policy Coordination Office at 301-763-6440."

1. Install Anaconda (https://www.anaconda.com/products/individual)
2. Install VS Code
3. Download Repository
4. See environment.yml file for dependencies
5. Run the primary Jupyter Notebook in the main folder

The housing unit inventory file names include the following convention:

• Acronym = hui - for housing unit inventory
• version = v2-0-0 - version of the file
• community = community name - name of the community ex. Lumberton, NC
• year = year of the data - ex. 2010
• random seed = random seed used to generate the data - ex. rs1000

Example description of the file name hui_v2-0-0_Lumberton_NC_2010_rs1000.csv:

• Version 2 Housing Unit Inventory for Lumberton, NC, using 2010 Census data. Set the random seed to 1000 to replicate the data.

Each data file is a comma separated value (CSV) file and each data file has a PDF file associated with it. The PDF file is the Codebook for the data file. The Codebook includes an overview of the project, details and notes on each variable, links to verify the data, and figures for exploring the data. NOTE - To understand the data please refer to the Codebook.

Program file names are designed to be unique and descriptive. The file names provide a way to identify the source of the data (based on the acronym), the order the programs should be run, and the version of the program. All program files (python files) are named with the following convention:

• Acronym (e.g ICD - for Intersect Community Data)
• Data science workflow step:
  • 00 - Administrative, Overview of Data, General Utility Functions
  • 01 - Obtain Data (read in ata from api, wget, data parsing, etc.)
  • 02 - Data Cleaning (merging files, adding new columns, checking keys, etc.)
  • 03 - Data Analysis
  • 04 - Data Visualization
  • 05 - Data Output (combined workflow steps for producing csv files)
  • 06 - Data Production (codebook, data sharing, archiving, etc.)
  • 07 - Data workflows (produce files from other programs)
  • 08 - Validation workflows
  • 09 - Uncertainty propagation workflows
• Program run order step (e.g run a before b or b depends on code in a)
• version (e.g v2 - version 2 of the file)
  • Github used for incremental version control, version number updated when program file has a major change.

The directory structure is as follows:

├───OutputData [ignored by github folder]
├───OutputData_LogFiles [ignored by github folder]
├───SampleOutputData [manually added to github repository to show example results]
└───pyncoda
    ├───CommunitySourceData
    │   ├───api_census_gov

The pyncoda directory contains the source code to replicate the data. In the pyncoda directory, the notebook file ncoda_06dv1_run_HUI_v2_workflow.ipynb is the notebook file that contains the workflow for the data replication.

The replication code creates the folder OutputData to store the data files and log files. This folder is not included in the GitHub repository or the DesignSafe-CI archive. The folder OutputData is ignored by github and was removed to limit the size of the DesignSafe-CI archive. The file directory_file_tree_2021-04-01.txt contains an archive of all files generated by the replication code.

The Archive directory contains the program files that were used to develop code but are no longer needed. These files are kept in case the code needs to be reused.

The folder pyncoda contains the source code to replicate the data - this source code archives the version of the code at the time the data was generated. A current version of the source code is available on GitHub at https://github.com/npr99/intersect-community-data/.

Within the pyncoda folder the folder CommunitySourceData contains the code to obtain, clean, and process the source data for a community. For the Housing Unit Inventory the source data is obtained from the Census Bureau API (apo.census.gov). The folder name api_census_gov reflects the url where the source data can be found - note that underscores (_) are used in place of . to allow for the folder to be used as part of the python package.

The folder OutputData_LogFiles is included in the DesignSafe archive but ignored by github. This directory contains the log files generated by the replication code. This folder was created manually to archive the details stored in the log files. The log files include all of the weblinks, and steps in the workflow.

[Work on acknowledgments] Mastura Safayet - PhD Student Texas A&M University [has fork of repository running in VS Code on local machine]