Can natal origins of biosecurity pests be determined using natural abundance isotopes?

Abstract

To know the likely natal origin of exotic pests is highly desirable for decision-making during biosecurity incursion responses. For example, a freshly arrived pest requires a different, much lower scale of response and resource than an established locally breeding population, risk pathways could be identified to inform pre-border protection biosecurity policy, and to distinguish individuals as vagrant rather than locally established can help verify a region's pest free status to maintain trade access. Natural abundance stable isotopes are intrinsically incorporated into tissues of insects via their food and water sources as they develop. These potential markers of natal origin have also been shown to vary at spatial scales useful for biosecurity. Specifically, multivariate analysis of isotope ratios that reflect differences in geology (e.g. 87Sr/86Sr) and elevation/climate (δ2H and δ18O), enabled accurate reassignment of 75% of individual insects to their geographic origin in a recent proof of principle study. Conversely, using just deuterium plus some knowledge of likely origin was used successfully to determine the source seaport of a biosecurity pest destined for Australia from New Zealand. Consequently, a project is underway to examine the validity of this technology for natal origin determination of pest fruit fly specimens. However, unlike more conventional isotope provenancing applications, the mass of biosecurity specimens and their numbers are very low, challenging the current technology and statistical confidence. Plus there is rarely knowledge of likely originating geographic area, food plant or environmental conditions. Therefore, beyond the useable geo-climatic variation, there could be a plethora of unknown biotic and abiotic influences on isotope ratio variation. To establish limitations, work is underway to assess generic principles such as: the applicability of tissue specific micro-sampling; the effect that different modes of feeding and natal host-plant may have on stable isotopes signal expression; the potential for models to measure and overcome such potentially confounding variation; and the statistical confidence of associated with small sample sizes when using multiple isotope ratios for provenance determination.