Wolf Poaching is not something happening
only in distant regions
Three sentence summary: Poaching is not something happening only in distant regions, it is the most common cause of wolf mortality in every population where it has been measured accurately. During the period U.S. wolves were listed under the ESA, the relative importance of poaching was systematically and substantially under-estimated while the relative importance of legal causes of mortality was systematically over-estimated. We correct the algebraic errors and errors of inference that led to these biased estimates.
Last year, Treves and others notified experts on Northern Rocky Mountain wolves that the assumption was flawed and in the ensuing year, the current team of authors investigated red wolves and Mexican wolves to confirm the same phenomenon applied. We now believe our finding applies to all studies of marked animals in which a perfectly reported cause of death occurs alongside imperfectly reported ones. Moreover, for populations with cryptic poaching – in which poachers conceal evidence – the biasing effect of the false assumption will be amplified.
Why does the article state that one error is a mathematical fact? Isn’t this a dispute over interpretation of data? We identified an error that is simply algebraic and an error of estimation that are separate issues. The algebraic mismeasurement is the over-estimation of the risk of legal killing. This should always be calculated as a proportion of all dead animals, not as a proportion of known fates because known fates over-represent legal causes of death by a known amount (Figures 1a,b below). We also identified an error of inference about the other causes of death. After one corrects the calculation of risk of legal causes of death as above, then one has to confront what might have happened to the unknown fates.
These unknown fates have been ignored traditionally, which discards useful information. We therefore presented a method to estimate what happened to those unknown fates. In the case of marked wolves, we show that poaching in particular has been under-estimated because cryptic poaching has not been accounted for properly. We presented two methods to account for cryptic poaching and one method that ignores cryptic poaching