Illegal international trade in wildlife (excluding fisheries and timber) has been valued at more than US$20 billion. A more precise figure has not been determined in part because of the clandestine nature of the trade, and for this same reason even regional and local levels of wildlife trade are difficult to assess. The application of recent developments in wildlife field-survey methods (e. g., occupancy) now allows for a more-accurate estimation of wildlife trade occurrence, including its hidden components at a variety of scales (e. g., regional, local) and periods (e.g., single season, 1 year, multiple years). Occupancy models have been applied in wildlife field studies to address the problem of false absences when conducting presence-absence surveys. Occupancy surveys differ from traditional presence-absence surveys because they incorporate repeat surveys, allowing for the likelihood of detecting a species (the probability of detection) to be estimated explicitly (in contrast to traditional surveys that often incorrectly treat this probability as close to one to allow for estimation of presence). Occupancy methods can be applied to a variety of wildlife-trade surveys, including, for example, single-species availability, links between two illegally traded species (i.e., co-occurrence), and disease occurrence in live trade. In addition, free user-friendly software (i.e., PRESENCE) allows even nonstatisticians to adequately address this issue. I simulated a hypothetical wildlife-trade market survey that resulted in an apparent 20% decline in naive occupancy (proportion of surveyed towns engaged in the trade) over 2 years, but when I accounted for change in probability of detection over the years the difference in occupancy was not statistically significant. As more sophisticated methods, such as occupancy, are applied to wildlife-trade market surveys, results will be more robust and defensible and therefore, theoretically, more powerful when presented to conservation policy and decision makers.
