Visualizing Wildfire Response: A Spatiotemporal Analysis of Social Media Photos Uploads in Los Angeles
dc.contributor.author | He, Yuxin | |
dc.contributor.advisor | Van Berkel, Derek | |
dc.date.accessioned | 2025-04-30T16:52:48Z | |
dc.date.issued | 2025-05 | |
dc.date.submitted | 2025-04 | |
dc.identifier.uri | https://hdl.handle.net/2027.42/196948 | |
dc.description.abstract | Climate change is intensifying the frequency and severity of extreme weather events, including wildfires, with profound impacts on landscapes, communities, and human behavior. Understanding public responses to such events is essential for improving emergency communication, preparedness, and recovery strategies. This study investigates behavioral changes associated with wildfires by analyzing over 140,000 geotagged Flickr images from the Los Angeles metropolitan area between 2005 and 2018. Using computer vision techniques and geospatial analysis, we classified visual content related to fire, smoke, burn scars, humans, and animals to examine how public photo-sharing behavior evolves before, during, and after wildfire events. Contrary to the initial hypothesis that wildfire activity would suppress public engagement, the results reveal a significant surge in social media photo activity surrounding fire ignition dates. Peaks in person-tagged images suggest that individuals often respond to wildfires not by withdrawing, but by documenting and sharing experiences in real time. However, explicit imagery of fire, smoke, and burn scars was relatively rare, highlighting a tendency toward human-centered storytelling over direct depictions of environmental destruction. Spatial analysis shows that fire-related photo clusters are concentrated in wildland-urban interface zones, and temporal trends confirm that engagement is highly event-driven and short-lived. While short-term public engagement increases, the study raises questions about long-term recreational use of fire-affected landscapes—a topic that warrants future longitudinal research. These findings demonstrate the value of leveraging social imagery as a real-time and spatially grounded lens for analyzing human responses to environmental disasters. They also underscore the need for improved computer vision models and sustained investigation into the longer-term behavioral impacts of climate-exacerbated wildfire regimes. | en_US |
dc.language.iso | en_US | en_US |
dc.subject | wildfire | en_US |
dc.subject | computer vision | en_US |
dc.subject | climate change | en_US |
dc.subject | social media | en_US |
dc.title | Visualizing Wildfire Response: A Spatiotemporal Analysis of Social Media Photos Uploads in Los Angeles | en_US |
dc.type | Thesis | en_US |
dc.description.thesisdegreename | Master of Science (MS) | en_US |
dc.description.thesisdegreediscipline | School for Environment and Sustainability | en_US |
dc.description.thesisdegreegrantor | University of Michigan | en_US |
dc.contributor.committeemember | Brines, Shannon | |
dc.identifier.uniqname | yuxih | en_US |
dc.description.bitstreamurl | http://deepblue.lib.umich.edu/bitstream/2027.42/196948/1/He_Yuxin_Thesis.pdf | |
dc.identifier.doi | https://dx.doi.org/10.7302/25446 | |
dc.description.mapping | d0a18e86-7d9e-4669-812b-ead353cc4899 | en_US |
dc.working.doi | 10.7302/25446 | en_US |
dc.owningcollname | Dissertations and Theses (Ph.D. and Master's) |
Files in this item
Remediation of Harmful Language
The University of Michigan Library aims to describe its collections in a way that respects the people and communities who create, use, and are represented in them. We encourage you to Contact Us anonymously if you encounter harmful or problematic language in catalog records or finding aids. More information about our policies and practices is available at Remediation of Harmful Language.
Accessibility
If you are unable to use this file in its current format, please select the Contact Us link and we can modify it to make it more accessible to you.