Determining Sublethal Ozone Stress Thresholds in Wild Pollinators

Overview

Ground-level ozone degrades floral volatile organic compounds (VOCs), potentially impairing pollinator navigation before population declines become detectable. However, no field-based study has examined how naturally varying ozone concentrations affect multiple wild pollinator taxa simultaneously in native chaparral ecosystems. This study investigated whether ambient ozone variation (AQI 35–115) across five Coastal Mountains chaparral sites correlates with changes in foraging behavior and pollination network stability among native bees, hoverflies, butterflies, and beetles. Twice-weekly field recordings were analyzed using computer vision and acoustic frequency analysis to quantify visitation rate, flight path entropy, velocity variance, and wingbeat variability. Bipartite pollination networks were constructed to assess density and species centrality across the ozone gradient. Breakpoint analysis identified and quantified a novel sublethal ozone stress threshold at AQI ~70, above which visitation rates declined, flight path entropy increased, and network density decreased nonlinearly. A Random Forest model integrating behavioral and network metrics accurately predicted pollinator stress probability, with SHAP analysis identifying visitation decline and entropy increase as dominant predictors. A composite Pollination Stress Index was developed as a field-validated early-warning metric for ozone-driven ecological destabilization. This study establishes the first integrated behavioral–network–AI framework for detecting sublethal atmospheric pollution stress in wild pollinator communities.

Competition history

• CSEF 2026 Earth & Environmental Sciences(Senior Division) · Entry S-08-09

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

• All Earth and Environmental Sciences projects
• CSEF Earth and Environmental Sciences projects
• Every CSEF 2026 project