Erratum for the Research Article “Invasion impacts in terrestrial ecosystems: Global patterns and predictors”

In response to an anonymous email sent to Science regarding the Research Article “Invasion impacts in terrestrial ecosystems: Global patterns and predictors” by M. P. Thakur et al. (23 October 2025, 10.1126/science.adq3101), the authors rechecked their data and identified eight studies that did not meet their non-native criteria because the focal invasive species were either native to the geographic region of the study or not clearly shown to be non-native there. As a result, they removed seven studies on native invasive (encroaching) plant species and one on a native invasive animal species from the analysis dataset, reducing plant studies from 512 to 505, animal studies from 93 to 92, and effect sizes from 1408 to 1385 (plants) and 295 to 294 (animals). All figures were updated; the only substantive change is in Fig. 3, where the breakpoint for soil moisture versus invasion minimum residence time shifted from 32 to 27 years. The authors also changed “AP” to “available P” in the legend of Fig. 3 to match the terminology in the legends of the other figures. Databases and species lists were revised accordingly. In the supplementary materials, the authors clarified that invasion duration was either derived directly from original studies or calculated from first-record and experimental years when not reported. They also clarified inclusion criteria for invasive non-native microorganisms. None of these updates alters the main conclusions of the study.