Real-time automated species level detection of trade document systems to reduce illegal wildlife trade and improve data quality
Biodiversity fuels the international wildlife trade, much of which is illegal and/or unsustainable. Border agents are typically overburdened, having to manually inspect import documents to ensure the legality of contents for often ≥100 shipments per day. Delaying shipments for inspection must be balanced against maintaining live animal welfare and reducing undue costs for traders. Biodiversity within the wildlife trade cannot be accurately estimated because of the multiple harmonization systems used to organize business transactions. Harmonizing wildlife trade data ignores species level classifications and aggregates data at less granular taxonomic or commodity level groupings. Here we describe a Real-Time Automated Species-Level Detection (RTASLD) system that assesses shipment declarations and invoices to collect data on species being traded. We use this to demonstrate how taxonomic imprecision on declarations and invoices can blur trade statistics and, at worst, be intentionally manipulated to conceal illegal wildlife. We address how taxonomic imprecision can interplay with Convention on the International Trade in Endangered Species of Wild Fauna and Flora (CITES) listed species. When only one or a subset of species within a genus are CITES listed, referred to as a Mixed-CITES-Genus, illegal trade can occur by identifying only to the genus level, and not to the species level which requires a declaration and CITES paperwork. RTASLD can be used to help border wildlife inspectors identify increased risk or presence of illegal wildlife trade that is occurring. Accurate species-level collection of trade data will help to better track biodiversity, stop illegal wildlife trade and maintain business expedience.
Tlusty, M., Cawthorn, D., Goodman, O., Rhyne, A. L., & Roberts, D. (2023). Real-time automated species level detection of trade document systems to reduce illegal wildlife trade and improve data quality. Biological Conservation, 281 https://doi.org/10.1016/j.biocon.2023.110022