Rapid assessment of disaster damage using social media activity

Abstract

Could social media data aid in disaster response and damage assessment? Countries face both an increasing frequency and an increasing intensity of natural disasters resulting from climate change. During such events, citizens turn to social media platforms for disaster-related communication and information. Social media improves situational awareness, facilitates dissemination of emergency information, enables early warning systems, and helps coordinate relief efforts. In addition, the spatiotemporal distribution of disaster-related messages helps with the real-time monitoring and assessment of the disaster itself. We present a multiscale analysis of Twitter activity before, during, and after Hurricane Sandy. We examine the online response of 50 metropolitan areas of the United States and find a strong relationship between proximity to Sandy's path and hurricane-related social media activity. We show that real and perceived threats, together with physical disaster effects, are directly observable through the intensity and composition of Twitter's message stream. We demonstrate that per-capita Twitter activity strongly correlates with the per-capita economic damage inflicted by the hurricane. We verify our findings for a wide range of disasters and suggest that massive online social networks can be used for rapid assessment of damage caused by a large-scale disaster.

Big data has big potential for applications to climate change adaptation

“…Researchers in the aftermath of Hurricane Sandy found the use of Twitter data to be more effective in predicting the location and severity of storm damage than models developed by the US government's Federal Emergency Management Agency (11). Bottom-up perspectives on adaptation development, implementation, and effectiveness can be documented from social media through sentiment analysis of large volumes of posts or by examining "Likes" abundance and distribution, ebird.org/content/ebird), crowdsourcing approaches can be expanded to gather and track information on adaptation policy on the ground.…”

“…Passively collected digital data have the potential to enhance the monitoring of climate-related threats and vulnerabilities, and can provide real-time awareness and feedback to decision makers and emergency services. Hazard warning systems, for example, could incorporate social media data to trigger emergency response measures (e.g., heat or flood alert systems); personal devices equipped with sensors could allow the monitoring of human movement before, during, and after a hazard event to aid with disaster response; tweets can be geotagged so disaster management services can map impacted areas in real-time to target efforts, and the Internet can be scraped for recently uploaded photos of affected areas (10,11). Search queries could be analyzed to monitor health-seeking behavior to detect outbreaks of climate-related diseases, and changes in the magnitude and frequency of climatic risks could be detected through time-series analysis of multiscale data, with the potential to detect "leading indicators" of abrupt, nonlinear change (12).…”

Employing Crowdsourced Geographic Information to Classify Land Cover with Spatial Clustering and Topic Model

“…Assuming that POI categories in one region are considered as words in one document, two outputs, the calculated topics (category distributions) and topic distributions, can be indicators of land cover classification. One of the output topic distributions Θ is shown in Equation (1): Two variables, which influence output distributions of POI categories and topics, are parameter α and topic number K. In our study, we employ the selection of value of α proposed by Griffiths et al [20]. To determine the proper topic number K, which is defined based on various categories of POIs, the perplexity algorithm is involved to evaluate the classified topics with specific topic number K. It refers to the uncertainty that a document belongs to one topic.…”

“…Several researchers have applied many CGI resources including geo-tagged photos [8, 9,16], check-in data [17], POIs [6,18], OSM [10,11] and other CGI. CGI has been used in a variety of applications, such as environmental detection [19], disaster management [20], urban land use identification [21], and land cover validation [4,22]. Since CGI is generated by different level of volunteers, data quality varies across space [12].…”

Public crowdsensing of heat waves by social media data

“…Among the different fields, scholars have begun to use SM to extract information about disaster events (Kryvasheyeu et al, 2016;Preis et al, 2013). On social media we may find digital traces of a disaster which can be used to derive the strength and impact of an event.…”