Two sides of the same coin: Does alien Impatiens balfourii fall into an ecological trap after releasing from enemies?

Abstract The evolution of increased competitive ability (EICA) hypothesis assumes that recently introduced populations of alien species are under low pressure from enemies, which allows them to reduce investment into defence and re-allocate the saved resources into the improvement of characteristics related to competitive abilities. We performed EICA tests to check if the low dispersal potential of Impatiens balfourii in Europe results from a limited evolution of its capability to release from enemies or from the low phenotypic plasticity of this trait. We tested habitat preferences of the species and compared the levels of enemy pressure in different habitats. This allowed us to determine if the tendency to occur along roadsides is manifested at the European scale in this species and if this preference is determined by a lower pressure of enemies in that habitat. It cannot be ruled out that occurrence along roads can be maladaptive for the species, and this habitat can potentially constitute an ecological trap due to frequent mowing. The study was carried out in six European populations differing in age and in five types of habitat. Enemy pressure was estimated by assessing leaf damage in adult plants and fungal pathogen load at the seed stage. We tested if the saved reserves are invested in the size and/or fertility of adult plants or in the seed quality. Activity of pollinators in different populations and habitats was also assessed. The majority of the results for adult plants indicate that the younger populations of I. balfourii release from enemy pressure better than the older ones. However, we found only weak support that the reserves previously spent on defence against natural enemies could be re-allocated into larger size or increased fertility. Only in the youngest population from Andorra was it largely confirmed. However, at seed stage, we found no support for the EICA assumptions in any of the populations. Thus, in general, the results of our study provide a fairly weak support for the EICA hypothesis. In comparisons between the habitats, we found that in five of the six populations, the species preferred roadsides, which was probably determined by the lower enemy pressure there. However, it cannot be excluded that this preference can be maladaptive and that this habitat can potentially constitute an ecological trap due to frequent mowing. Notably, the species was also abundant along streams, where the plants are less disturbed by human activities. This finding was particularly pronounced at the invasion forefront in Andorra, where the species was completely absent from roadsides. Therefore, we assume that unless control measures are undertaken, streamsides may be more important for the future invasion of I. balfourii than roadsides in that area. The ornamental value of I. balfourii makes it an obvious candidate to replace its banned invasive counterpart I. glandulifera in trade. The increase in popularity in trade and cultivation would inevitably lead to an increase in propagule pressure, which could be crucial for increasing the rate of invasion of I. balfourii on a large scale in Europe.