When Disaster Strikes
How data visualization can aid in disaster relief.
Researchers from the University of Colorado, Boulder are developing a new data visualization tool that will aid in the integration of data to improve large-scale disaster operations and co-located teamwork across various industries.
Crisis informatics, also referred to as “disaster informatics,” has become a pivotal field of research bridging information technology, social sciences, and emergency response experts to help those impacted by disasters and ultimately improve preventative practices. The ways in which society handles crises has changed drastically, from the rise in social media to the technologies used by field experts to alleviate these disasters. From wildfires rampant in the state of California to the continually looming presence of the coronavirus, it’s imperative that crisis teams are equipped with the tools to help inform quick and effective decision-making and minimize errors that can prove to be dire.
Data visualizations have the power to transform practices across many industries, enhancing the ways analysts collect, interpret, and execute data to enhance decision-making. Current practices for fieldwork require analysts to pre plan their operations based on data previously collected, then using mobile devices or in field notebooks, physically transport new data to a central location to synchronize with other sources. In many instances, “when” and “how” that data is collected can have serious repercussions if it’s not utilized efficiently. For example, if an aerial firefighter cannot accurately determine the geolocations of the fires and/or nearby communities, it will take much longer to alleviate the fire and could result in serious injuries.
“Situational awareness (SA)” is also an important concept used by analysts to perceive, comprehend, and identify data elements relevant to the environment they’re working in — essentially, knowing what’s going on around them. SA is a critical aspect when discussing field data analysis and how it affects crisis teams ability to detect changes in their environment in real-time, prepare for unexpected interruptions, reorienting their tasks. Increasing SA is especially critical in technologies for highly dynamic situations where new data frequently changes operational strategies, such as search-and-rescue and emergency response.
Researchers at the University of Colorado, Boulder (UC) conducted a series of case studies and interviews to explore and understand how data visualization tools can be designed to effectively support the needs of field data analysts and stimulate conversations around a deeper data integration into their workflow. During the interviews with analysts from a diverse set of industries, such as geological sciences and public safety, the researchers were able to pinpoint key limitations that prevent analysts from better leveraging the data out in the field:
1.) Divided attention between navigating the data and the environment they’re immersed in, as well as overcoming information overload
2.) Smaller versions of devices, known as small form-factors, used outside of field work, i.e. mobile and tablet devices
3.) Touchscreen input, which can significantly hinder those who wear protective gloves or other hand coverings
4.) Reliance on operation centers to provide and update critical data.
To address these issues, the research team designed and developed a data visualization tool called “FieldView.” FieldView is an open-source prototype that incorporates augmented reality visualizations to improve colocated teamwork, data validation and refinement, and integrates automated sensors which can help stream data faster and from more locations, even without a powerful internet connection, by using a portable server.
Incomplete and poorly collected field data costs analysts time and money, but is difficult to detect while in the field despite analysts being well-positioned to correct the errors. By developing FieldView and using data visualizations in the context of the operational area, field analysts can identify and fix these errors as they occur in real-time, leading to substantial savings in both time and budget. The UC researchers are continuing to work towards providing more formative insight into how visualizations can enhance field work using FieldView as a first, critical step.
Analysts can also use FieldView to explore patterns across their environment to look for inconsistencies that might suggest outdated data. Once those inconsistencies are located, analysts can walk directly to any problematic areas, assess the source of the inconsistency and update data.
Data-oriented decision making is transforming practices in crisis response and a broad variety of additional industries, including earth science, geology, and public safety. By developing tools like FieldView to better leverage field data to describe the state of these complex environments and make more informed decisions, we can begin to improve future data analysis practices that will improve the overall quality of life for all.
To read more about the FieldView project, visit the University of Colorado’s College of Media, Communication and Information website.
This story is based on the following paper:
- Matt Whitlock, Keke Wu, Danielle Albers Szafir. Designing for Mobile and Immersive Visual Analytics in the Field. IEEE Transactions on Visualization and Computer Graphics, 2019.
