Systematic Literature Review on Merging AI-Based Wildfire Detection with Bee Bioacoustics: A Hybrid Environmental Sensing Approach

Source abstract

The increasing frequency and severity of wildfires, exacerbated by climate variability and human activities, demand innovative solutions for early detection and risk assessment. This systematic review critically examines the convergence of advanced wildfire prediction technologies, including machine learning, satellite remote sensing, and IoT sensor networks, with bees’ behavioural and physiological responses to environmental stressors. Special emphasis is placed on the emerging potential of bee acoustic monitoring as a non-invasive, nature-inspired method for detecting subtle environmental changes that may precede wildfire events. By synthesizing findings from over 200 peer-reviewed articles published in recent years, this review identifies key environmental parameters, like temperature, humidity, smoke, and CO2 that influence both wildfire dynamics and bee colony behaviour. The analysis highlights both the promise and challenges of integrating AI-driven systems with bioindicator species like bees, including issues of data quality, model generalisation, and multi-modal data fusion. Ultimately, this review underscores the value of a multidisciplinary, bio-inspired approach to wildfire early warning systems and outlines future research directions to enhance the accuracy and robustness of wildfire detection frameworks.