Bayesian Probability Modelling Approach Research Articles

Absence of evidence is not evidence of absence: Feral pigs as vectors of soil‐borne pathogens

AbstractInvasive soil‐borne pathogens are a major threat to forest ecosystems worldwide. The newly discovered soil pathogen, Phytophthora ‘taxon Agathis’ (PTA), is a serious threat to endemic kauri (Agathis australis: Araucariaceae) in New Zealand. This study examined the potential for feral pigs to act as vectors of PTA. We investigated whether snouts and trotters of feral pigs carry soil contaminated with PTA, and using these results determined the probability that feral pigs act as a vector. We screened the soil on trotters and snouts from 457 pigs for PTA using various baiting techniques and molecular testing. This study detected 19 species of plant pathogens in the soil on pig trotters and snouts, including a different Phytophthora species (Phytophthora cinnamomi). However, no PTA was isolated from the samples. A positive control experiment showed a test sensitivity of 0–3% for the baiting methods and the data obtained were used in a Bayesian probability modelling approach. This showed a posterior probability of 35–90% (dependent on test sensitivity scores and design prevalence) that pigs do vector PTA and estimated that a sample size of over 1000 trotters would be required to prove a negative result. We conclude that feral pigs cannot be ruled out as a vector of soil‐based plant pathogens and that there is still a high probability that feral pigs do vector PTA, despite our negative results. We also highlight the need to develop a more sensitive test for PTA in small soil samples associated with pigs due to unreliable detection rates using the current method.