Abstract ID: 3.12924 | Accepted as Talk | Poster | TBA | TBA

Mountain regions are fragile ecosystems and biodiversity hotspots providing important ecosystem services, yet they face increasing threats from climate change and human activities. Litter—misplaced solid waste such as plastics, metal, paper, and organic materials—poses a significant challenge. However, high-altitude regions remain understudied in this context, with most research offering spatially constrained insights. Comprehensive data on litter distribution are urgently needed to mitigate ecological impacts and develop effective management strategies. Here, we combined smartphone-based citizen science by the usage of the litter app “Dreckspotz” with deep learning to address data quality challenges and analyze litter distribution across Austria over 7 years, with a focus on alpine regions. We employed deep learning models to effectively remove low quality data, enhancing data reliability and strengthening study conclusions. Our analysis revealed that plastics and cigarettes dominated litter categories, and the five most frequently reported brands accounted for 75% of the reported litter, emphasizing the role of specific products in environmental pollution with altitudinal variations. These findings highlight the potential of citizen science to bridge data gaps in challenging environments and demonstrate the value of advanced validation techniques to improve data quality. By identifying environmental pressure points, this study provides actionable insights to guide conservation efforts and inform policies aimed at preserving fragile mountain ecosystems.