1. Many invasion hypotheses propose biotic interactions as the main mechanism to explain non-native species' success.  Despite the evidence that the strength of biotic interactions varies with abiotic context, it remains unclear whether the importance of the different mechanisms proposed to explain invasion predictably varies with abiotic context and whether this variation is consistent across different growth forms.

2. We reviewed studies at a global scale to evaluate whether evapotranspiration, latitude, precipitation, and temperature influence the importance of disturbance, enemy release, facilitation, and novel weapons mechanisms to explain non-native plant invasions.  In total, we calculated 171 effect sizes for ~300 non-native plant species covering a wide range of environmental conditions and growth forms.

3. Environmental context and plant growth form influenced the role played by each invasion mechanism.  The importance of disturbance in facilitating invasion exhibited a quadratic relationship with latitude and temperature and decreased with increasing precipitation.  In mixed communities and trees, disturbance was mediated by either evapotranspiration, latitude, precipitation, or temperature.  Enemy release exhibited a quadratic relationship with evapotranspiration, latitude, and precipitation, and was positively related to temperature.  The importance of enemy release was also contingent on growth form and was highly context-dependent, enemy release responses for grasses, and trees were modulated by either evapotranspiration, latitude, precipitation, or temperature.  The importance of facilitation decreased with increasing temperature.  In forbs, facilitation decreased with evapotranspiration and temperature.  The importance of novel weapons was more strongly confirmed for studies conducted at lower evapotranspiration, precipitation, and higher latitudes, and exhibited a quadratic relationship with temperature.

4. Synthesis. Our results show that environmental conditions not only filter non-native species depending on physiological tolerances but may also influence the importance of invasion mechanisms.

Dataset used in meta-analyses and meta-regression models in Chiuffo et al. (2022). Columns A-G contain study and invasion mechanisms information. Columns H-W contains effect sizes and variance estimates. Full citations for each publication can be found in Appendix S4. Columns X-AA contains the four explanatory predictor variables used.