NSF Award
2430277
Biodiversity loss is one of three critical interlinked challenges facing humanity today, alongside climate change and pollution. Nature crime, particularly, undermines the effectiveness of activities to reduce biodiversity loss. Among these crimes, illegal wildlife trade (IWT) is egregious and a cause and consequence of biodiversity loss. This illegal activity spans all US states and territories. The proposed research will use new science-practitioner partnerships to overcome scientific knowledge failures about a) data generation and classification and b) data integration and application. The science team will use sharks, rays, and turtles as conservation examples and include expertise in molecular biology, wildlife forensics, operations research, network disruption, computer and data science, conservation criminology and human geography. This research will enhance conservation efforts both in the US and globally by increasing scientific awareness and improving the precision of knowledge about the scope and rate of loss of sharks, rays and turtles involved in the illegal wildlife trade. Findings will contribute valuable data about these species and will also be integrated with other data to better understand fundamental changes of socio-environmental systems. Additionally, the results will inform science-based strategies for disrupting illegal wildlife trade including crime prevention, restorative justice, and law enforcement measures. The science team will also engage with undergraduate students through project-based learning, support PhD dissertations, and provide specialized training for law enforcement and their partners through a 5-day tool-training workshop. Furthermore, the researchers will collaborate with diverse stakeholders by sharing information, co-designing initiatives, and offering decision-making support.<br/><br/>The research aims focus on novel species identification technologies, online market analysis, data integration, and operational strategies to address fundamental challenges hindering effective action for IWT. The goals and scope of this research include: 1) producing near real time species-level genetic identification for 185 species using unique High-Resolution Melt (HRM) profiles; 2) designing and developing new machine learning frameworks that explore various HRM ranges, utilizing advanced deep learning and transfer learning approaches using data augmentation techniques; 3) advancing the accuracy of species identification through improved analysis of HRM curve profiles; 4) conducting large-scale data collection, innovative data labeling, and automatic classification to provide data openness, high recall rate, effective IWT post classification models, and a visualization tool; 5) integrating physical and virtual crime ecosystems using spatially interoperable data from experts and non-experts; 6) addressing adversarial challenges through adaptive learning and sequence creation to improve decision-making under uncertainty; and 7) developing a model structure capable of accounting for complexity of real-world networks. This research will advance science understanding and help overcome conservation knowledge failures, thereby aiding efforts to decrease the acceleration of biodiversity loss from IWT.<br/><br/>This project is jointly funded by the Divisions of Environmental Biology and Integrative Organismal Systems through the Partnership to Advance Conservation Science and Practice Program.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
