Conclusions
Extensive nationwide range expansion through secondary introduction is likely to be restricted by currently unsuitable climate beyond southern regions of the UK. However, exponential growth of local populations in habitats providing transport pathways (i.e., movement of aggregates, timber, plants, general public) is likely to increase opportunities for regional expansion. A similar scenario has been described in the spread dynamics of invasive Pallas’s squirrel (Callosciurus erythraeus ), where a constant increase in the appearance of new populations occurred after a two decade lag, and could be explained by increased vector activity (intentional translocations) as the population size at the initial introduction foci increased – causing a potentially exponential increase of translocation events when translocated populations start acting as a source themselves (Guichon, Benitez, Gozzi, Hertzriken, & Borgnia, 2015).
Despite the fundamental physiological importance of sun exposure to diurnal reptiles, to the best of our knowledge, our models are the first to incorporate estimates of solar insolation into models of probability of occurrence at this fine scale (but see Bennie, Huntley, Wiltshire, Hill, and Baxter (2008) for a detailed discussion on use of solar influx as a variable in modelling fine scale topographic microclimate). Our models demonstrate the inclusion of the variable at this scale, and indeed our entire approach to developing a fine scale SDM, could be very useful in other applications relating to ectotherm ecology (e.g., in developing habitat suitability indices, directing habitat management, guiding survey effort for rare/cryptic species). Furthermore, whilst the use of SDMs and IBMs have become a widely used to further understanding of mechanisms driving invasion dynamics (Kadoya & Washitani, 2010; Fraser et al., 2015; Suzuki-Ohno et al., 2017), the benefits of incorporating spatially explicit individual based models into management plans for the control of invasive species has only recently been recognised (Day, Landguth, Bearlin, Holden, & Whiteley, 2018). In this regard, our models provide a best estimate for future expansion of P. muralis at both the UK national and local scale, providing essential information (i.e., dispersal patterns, key habitat, current and projected population sizes) on which management decisions could be made. As our models show UK populations may be approaching an end to an inherent lag phase, there is argument to suggest that a timely precautionary intervention may be justified to halt some populations before an abrupt end to the lag phase occurs.
Data accessibility: the following data is in the process of being deposited with EIDC
-MaxEnt (local and national scale) input files including presence and background samples with climate/environment data
- Inputs for Rangeshifter models and subsequent population outputs