HEvOD: The First Database of Hurricane Evacuation Orders in the United States

A comprehensive database of historical hurricane evacuation orders in the United States for researchers, practitioners, and policymakers.
Like

Share this post

Choose a social network to share with, or copy the URL to share elsewhere

This is a representation of how your post may appear on social media. The actual post will vary between social networks

Are hurricane evacuation orders effective? To what extent does a hurricane evacuation order increase evacuation? Do people who live in different evacuation zones respond differently to evacuation orders?  These are important questions that can provide valuable insights to researchers, policymakers, and decision-makers to design and develop effective and life-saving evacuation strategies in anticipation or in response to hurricanes. While researchers, over the past three decades, have advanced our knowledge and understanding of how people make evacuation decisions during hurricanes (and in general, natural disasters), these questions are still left largely unexplored due in large part to the lack of a comprehensive, standardized, and high-spatial and -temporal resolution dataset of historical hurricane evacuation orders. That motivated us to develop the Hurricane Evacuation Order Database, or as we call it HEvOD.

HEvOD is a database of hurricane evacuation orders that were issued in the United States between 2014 and 2022. The database features evacuation orders that were systematically retrieved and compiled from a wide range of resources, including official websites and social media accounts of local and state governments and government agencies, as well as different news media platforms. The database includes information on the type of the order, the announcement date and time of the order, the effective date and time of the order, and the evacuation area, all systematically organized to facilitate easy access and analysis.

Compiling such a comprehensive database was ambitious and challenging. One of the primary hurdles we faced was navigating the disparate evacuation laws and policies across different states, coupled with the diverse platforms used for disseminating evacuation orders. The data collection process required meticulous planning and execution, relying on various sources including official websites, social media, and news platforms to ensure accuracy and comprehensiveness. Through a combination of manual searches and by utilizing social media API, the team managed to create a consistent, reliable dataset. Figure 1 shows some examples of how we extracted the relevant attributes and information for our database from different sources.

An example of how different attributes were extracted from different sources: (a) government postings, (b) Facebook posts, (c) news articles, and (d) Twitter posts.
Figure 1: An example from Hurricane Irma of how attributes were extracted from varied sources: government postings (a) detail the state of emergency declaration via executive order, Facebook posts (b) show a voluntary evacuation order issued by Indian River County, news articles (c) provide supplemental information when official posts are unavailable, highlighting Brevard County’s mandatory evacuation order from WFTV article, and Twitter posts (d) highlight Sarasota County’s mandatory evacuation orders. Attributes are color-boxed for clarity: order specifics in orange, types of evacuation orders in green, regions affected in red, and effective dates in blue.

The rich collection of attributes and the high-temporal resolution of data in our database, combined with other datasets (e.g., GPS-based mobility data, hurricane evacuation zones), enable detailed investigations into the effectiveness of different evacuation policies and allow researchers to causally understand the relationship these evacuation policies and evacuation decision-making. Figure 2 presents the spatial distribution of the evacuation orders extracted from the database, showing seventeen counties across Mississippi, Louisiana, Florida, South Carolina, and North Carolina issued at least four mandatory orders between 2014–2022.

Mandatory evacuation orders count.
Figure 2: The total number of mandatory evacuation orders issued by each county for the hurricanes in the HEvOD between 2014 and 2022.

HEvOD is hosted on LibraData,1 University of Virginia’s open-access institutional repository, and our website, www.hurrevacorder.info, provides a platform for interested users to execute queries to download the data and visualize mandatory evacuation orders. Looking ahead, our goal is to continuously update the database shortly after hurricane evacuation orders are issued in response to hurricanes in the future. We also welcome users to contribute to HEvOD using our website by filling out some of the gaps and missing data. We are excited about the opportunities that this database can offer to researchers to better understand the evacuation decision-making process and to policymakers and emergency management officials to better evaluate the effectiveness of evacuation policies with the ultimate goal of minimizing the risk of damage from hurricanes in the future.

1Anand, Harsh; Alemazkoor, Negin; Shafiee-Jood, Majid, 2023, "HEvOD: Hurricane Evacuation Order Database", https://doi.org/10.18130/V3/ZGS4T4, University of Virginia Dataverse, V4

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in

Follow the Topic

Natural Hazards
Physical Sciences > Earth and Environmental Sciences > Earth Sciences > Natural Hazards
Research Data
Research Communities > Community > Research Data
Social Policy
Humanities and Social Sciences > Society > Social Policy
Sustainability
Physical Sciences > Earth and Environmental Sciences > Environmental Sciences > Sustainability

Related Collections

With collections, you can get published faster and increase your visibility.

Epidemiological data

This Collection presents a series of articles describing epidemiological datasets spanning diverse populations, ecosystems, and disease contexts. Data are presented without hypotheses or significant analyses, and can be derived from population surveys, health registries, electronic health records, field sampling, or other sources.

Publishing Model: Open Access

Deadline: Mar 27, 2025

Neuroscience data to understand human behaviour

This Collection presents descriptions of datasets combining brain imaging or neurophysiological data performed alongside real-world tasks or exposure to different stimuli.

Publishing Model: Open Access

Deadline: Jan 30, 2025