Simulation check and uncertainty in introduction and naturalisation dates
I used simulation to check the methods described above worked as expected, using known parameter values to generate artificial data similar to the real data, fitting the lag time model to the artificial data and estimating the invasion debt, and assessing how well the methods recovered the true parameter values using repeated simulations. Full details are in Appendix S4.
The dates of first introduction and naturalisation in this study are taken from historical records and are subject to uncertainty, notably because the true dates might not have been documented or the records have not been uncovered, resulting in first record dates being more recent than true dates. While dates can be updated as new records are discovered, any set of dates will be subject to uncertainty due to the difficulty of determining precisely when past events occurred. The question is whether this uncertainty could affect the results reported in this study. To evaluate this, I used artificial data simulation as described above but added measurement error to the simulated dates. I then assessed how well the methods recovered the true parameter values in the presence of dating errors using repeated simulations. Full details are in Appendix S5.