Predicting the temperature-dependent natural population expansion of the western corn rootworm, Diabrotica virgifera

The western corn rootworm, Diabrotica virgifera virgifera Leconte (Coleoptera: Chrysomelidae), was accidentally introduced near Belgrade in Serbia just before 1992 and from there its expansion into Europe started. We have estimated its mean rate of expansion from 1992 to 2000 to be approximately 33 km year^?1, using data from the annual surveys of the European and Mediterranean Plant Protection Organisation. We investigated whether or not D. virgifera can establish itself at certain places in Europe, taking its temperature‐dependent development into account. We also estimated the time it will take D. virgifera to reach the Netherlands, considering only its dispersal by flight. All life stage transitions of D. virgifera were simulated with the program INSIM to assess whether or not it could establish itself in particular places. In the simulations, we used the available laboratory data on its life history characteristics, as well as a time series (2–6 years) of daily minimum and maximum temperatures from weather stations across Europe. The temperature‐dependent net reproduction resulting from the simulations showed that D. virgifera populations cannot establish themselves at latitudes above 55°N. The overall expected velocity of D. virgifera's range expansion was computed with van den Bosch et al.'s formula [ van den Bosch F, Hengeveld R & Metz JAJ (1992) Analysing the velocity of animal range expansion. Journal of Biogeography 19: 135–150], where the expansion velocity was based on dispersal characteristics and demographic parameters. We predicted that D. virgifera will not reach the Netherlands by flight from the border of its 2000 range before 2018 using this overall expected velocity. The explanation of this late arrival is that the velocity of population expansion decreases in all directions from the centre of its 2000 range due to unfavourable temperatures. Our analysis is an improvement on former analyses in that it uses temperature‐dependent life history characteristics. We stress, however, that the lack of knowledge on dispersal behaviour and on the values of life history characteristics in field situations might severely limit the applicability of the predicted velocity.