Automated Insect Detection and Biomass Monitoring via AI and Electrical Field Sensor Technology

Insects, vital for ecosystem stability, are declining globally, necessitating improved monitoring methods. Traditional approaches are labor-intensive, invasive, and limited in scope. This study presents a novel, automated, non-invasive insect monitoring system that detects electrical field modulations caused by flying insects. Utilizing in-field sensors, the system measures activity and biomass without physical trapping. It employs differential electric field measurements and convolutional neural networks for insect detection and wing-beat frequency analysis, along with a biomass algorithm estimating species-specific weights. Field validation within 2 sites at a Danish nature reserve showed moderate to strong correlations between sensor and Malaise trap measurements, particularly at site 1 (Spearman’s $\rho=0.725$ for counts, $0.644$ for biomass), validating the potential of our method. Additionally, sensor-sensor correlations ($\rho=0.758$ for counts, $0.867$ for biomass) exceeded Malaise-Malaise correlations ($\rho=0.597$ for counts, $0.641$ for biomass), suggesting greater measurement consistency for the sensors. However, these differences were not statistically significant ($P=0.304$ for counts, $P=0.057$ for biomass). While the biomass $P$-value did not reach the conventional significance threshold ($P<0.05$), its proximity suggests a more stable biomass estimate than Malaise trapping. Overall, this innovative approach bridges critical gaps in insect monitoring, offering scalable, ethical, and efficient solutions for insect conservation and ecosystem management.