Crowdsourced and Social Media Data for Environmental Monitoring and Disaster Response: A Dual Case Study of Real-Time Flood Tracking and Air Quality Surveillance

The proliferation of information via mass media and social media has revolutionized the monitoring and management of natural and environmental disasters. Citizen science programs such as eBird and iNaturalist have become key to environmental monitoring, and social networks such as Twitter and Facebook are flooded with instant information. These platforms provide timely and spatially relevant information on events such asfloods, earthquakes, and wildfires, and are essential to data collection processes such as satellite imagery and weather sensors. Using natural language processing (NLP) and machine learning, data from these sources can be quickly analyzed to provide better outcome insights for law enforcement and emergency medical agencies. However, data quality, verification, and privacy issues remain, as user-generated content can be subject to misinformation and bias. Despite these challenges, new solutions, such as AI-enabled data access methods and the integration of crowdsourced data into systems, hold the promise of enabling early warning capacity and helping to monitor changes in biodiversity and air quality in areas lacking good infrastructure. This review includes the results of 22 studies that provide a broad overview of the role of crowdsourcing and community education in improving disaster preparedness, protection, and environmental monitoring, and that address key questions and make recommendations for future