The unrivaled growth in e-commerce of animals and plants presents an unprecedented opportunity to monitor wildlife trade to inform conservation, biosecurity, and law enforcement. Using the internet to quantify the scale of the wildlife trade (volume and frequency) is a relatively recent and rapidly developing approach that lacks an accessible framework for locating relevant websites and collecting data. We produced an accessible guide for internet-based wildlife trade surveillance. We detailed a repeatable method involving a systematic internet search, with search engines, to locate relevant websites and content. For data collection, we highlight web-scraping technology as an efficient way to collect data in an automated fashion at regularly timed intervals. Our guide is applicable to the multitude of trade-based contexts because researchers can tailor search keywords for specific taxa or derived products and locations of interest. We provide information for working with the diversity of websites used in wildlife trade. For example, to locate relevant content on social media (e.g., posts or groups), each social media platform should be examined individually via the site's internal search engine. A key advantage of using the internet to study wildlife trade is the relative ease of access to an increasing amount of trade-related data. However, not all wildlife trade occurs online and it may occur on unobservable sections of the internet.

The unrivalled growth in e-commerce of animals and plants presents an unprecedented opportunity to monitor wildlife trade to inform conservation, biosecurity, and law enforcement efforts. Using the Internet to quantify the scale of the wildlife trade (volume, frequency) is a relatively recent and rapidly developing approach, which currently lacks an accessible framework for finding relevant websites and collecting data. Here, we present an accessible guide for internet-based wildlife trade surveillance, which uses a systematic method to automate data collection from relevant websites. The guide is easily adaptable to the multitude of trade-based contexts including different focal taxa or derived parts, and locations of interest. Furthermore, as wildlife trade on the Internet becomes more widespread, the ability to collect large amounts of data on traded wildlife will become possible and desirable. Using a case study where we monitor 53 websites, we demonstrate the capabilities and limitations of this kind of large-scale surveillance system. We collected over half a million unique listings in a year and estimate that it would take over two years for one person to clean every listing. We propose that the development of machine learning methods for automation of data collection and processing become a priority and be tested for a variety of different contexts of wildlife trade-related web data.

No abstract