Abiotic factors
The variables representing abiotic environment (A) are the most important for model performance for both species; however, the impact of these variables was more pronounced in the case of S. gigantea , compared with S. canadensis .
The distribution of both species was restricted climatically, and their presence was unlikely in areas with an average yearly temperature below approximately 5.5°C. The temperature corresponds with the altitudinal zonation of vegetation in the studied region and relates to a lower limit of the montane zone, starting from an altitude of approximately 600–850 m a.s.l. in the studied region. The negative effect of cold climate on the distribution of both Solidago species studied is in accordance with studies examining their potential distribution in Europe, which indicated that northern Europe as a region is outside their climatic requirements (Weber, 2001). Although both species can be observed sporadically at higher altitudes, their typical upper limit is 1200 m a.s.l. (Moran et al., 2017; Weber & Jakobs, 2005). In the case of S. gigantea , positive correlations have been found between the mean temperature and growth parameters, and high spring temperatures (above 24 °C) are advantageous for germination (for review, see Weber & Jakobs, 2005). Solidago canadensis plants are taller at lower attitudes, and at higher altitudes, they are not able to develop seeds because of the limited length of the vegetation period (Moran et al., 2017). It should be noted that the data referred to here regarding altitude come from the central Alps, while the climate in the Carpathian Mountains is more severe; therefore, the upper limits of the vegetation zones are at lower altitudes in the Carpathian Mountains comparing with the Alps (Ellenberg, 1988; Pawłowski, 1972).
The species distributions were also correlated with temperature seasonality, which in the studied region is also related to the precipitation pattern (Appendix, Table S.3.). Solidago canadensisis more abundant in the western part of the study region, which has lower temperature seasonality and higher precipitation, while S. gigantea avoids the southern part of the region with higher precipitation and also lower temperature seasonality. Previous studies examining the potential range of this species in Europe (Weber, 2001) suggested that these aspects (continentality gradient and precipitation) did not restrict their distribution in this part of Europe. Therefore, the extent to which the observed relation is causal is not clear, and the possibility exists that it reflects a peculiarity of the distribution in the studied region.
The models did not indicate that soil properties and landrelief features are among the crucial factors explaining the distributions of the invaders. Both species are known to have rather broad tolerance to soils (Szymura & Szymura 2016; Weber & Jakobs, 2005; Werner et al., 1980), which could explain why soil properties were not relevant in studied region. Observations from early phase of invasion on studied region, up to 1989s, underlined the role of river valleys, as a main route of invasion (Tokarska-Guzik, 2005). The results obtained here show that the species are broadly widespread and their invasion is no longer related to watercourses.