Project Grant 2240402

This National Science Foundation (NSF) Project Grant award under the Social, Behavioral, and Economic Sciences (CFDA 47.075) program provides $250,000 to Virginia Polytechnic Institute & State University (Virginia Tech) to develop new machine learning approaches for enhancing the accuracy and resolution of stable isotope ratio analysis (SIRA) to trace the geographic origin of timber products. The objective is to improve the ability to detect and combat illegal trafficking and trade of natural resources. The 4-year project will combine expertise from data science, analytical chemistry, geospatial analysis, and international trade specialists to create new methods for overcoming limitations in labeled dataset availability. Key deliverables include developing contrastive learning, generative learning, and science-guided machine learning algorithms to integrate climate data and site-specific observations. The project will also explore active sampling strategies to quantify model uncertainty and optimize future data collection. The research aims to advance national capabilities for rapidly identifying the source of suspicious timber products, helping enforce international regulations and combat resource trafficking.