Data and code for the manuscript 'Schreiner et al. (2026). Slow and context-dependent: Responses to grassland extensification across a comprehensive set of ecosystem attributes.'

This supporting information accompanies the manuscript Schreiner et al. (2026), “Slow and context-dependent: Responses to grassland extensification across a comprehensive set of ecosystem attributes.” In this manuscript, we assess the potential of extensification as a grassland restoration strategy using a large-scale field experiment conducted across 45 grassland sites in three German regions. Sampling was conducted in 2023, after three years of experimental treatments. We evaluate above- and belowground responses to land-use reduction relative to predefined restoration targets and demonstrate that extensification induces rapid changes in many plant-related indicators, whereas belowground properties exhibit delayed responses. Restoration outcomes were strongly context-dependent, with greater progress observed in less productive regions, highlighting the importance of site-specific planning and long-term monitoring in grassland restoration.<br><br>Data and code are provided in a structured folder system that enables full reproducibility of all statistical analyses, figures, and tables presented in the manuscript.<br><br>The folder system has the following structure:<br><br>\analysis\data\Data.csv = the dataset containing all variables.<br><br>\analysis\data\Metadata.xlsx = the corresponding metadata.<br><br>\analysis\scripts\Effect_of_extensification_on_restoration_goals.Rmd = a R Markdown file used to conduct the analyses and generate outputs (stored in \analysis\output).<br><br>\analysis\scripts\Effect_of_extensification_on_restoration_goals.html = the corresponding HTML document. Figures and tables generated by the script are embedded in the HTML document.<br><br>\analysis\output\results\Effect_table.xlsx = a table summarising extensification effects for all variables (generated by the script).<br><br>\analysis\output\results\Model_table.xlsx = a table with parameter for all fitted models (generated by the script).<br><br>\analysis\output = a folder containing several subfolders in which residual diagnostic plots (generated by the script) for all fitted models are stored.