Plant extirpation at the site scale: implications for eradication programmes

Aim To identify the relative importance of commonly available site and species factors on the probability of extirpating a plant species at the site scale. Location Australia. Methods We reviewed the plant eradication literature and listed the factors commonly cited as influencing eradication success. We then analysed a database of 408 site-scale extirpation attempts, across 17 species eradication programmes, by modelling the time until extirpation to determine the relative influence of these factors on the probability of extirpation, given time. Results The most commonly cited factors influencing eradication success can be classified into two groups as follows: those related to the 'organisation', being typically binary variables and amenable to management; and those related to the 'site/species', which typically were continuous values and usually beyond the control of the management agency. Detectability period, search distance, monitoring rate, infestation size, propagule longevity, time to reproductive maturity and previous eradication success all influenced the extirpation rate. Conversely, climate suitability, land use and general accessibility (distance to nearest manager) were relatively unimportant. By relating the influential site/species factors to a time-dependent model, managers and policymakers can explicitly estimate the probability of successful extirpation at a site, given a particular time horizon. These estimates can then be aggregated up to the species scale to allow managers to set realistic goals regarding eradication time frames and resource requirements. Main conclusions Our paper illustrates how the probability of extirpation at the site scale can be quantitatively estimated and how these estimates can be used to inform decisions regarding broader species-scale eradication programmes. We expect that the use of such quantitative approaches to select better eradication targets will improve eradication success rates over time.