Read me file for dataset titled "20210128 remnant meta model summary all types and levels subset random.csv" Paper title: "Large ecosystem-scale effects of restoration fail to mitigate impacts of land-use legacies" Authors: Lars A. Brudvig (brudvig@msu.edu) Nash E Turley (nashuagoats@gmail.com) Savannah L. Bartel Lukas Bell-Dereske Sabrie Breland Ellen I. Damschen Sarah E. Evans Jason Gibbs Philip G. Hahn Rufus Isaacs Joseph A. Ledvina John L. Orrock Quinn Sorenson John D. Stuhler Author of this document: Nash E Turley (nashuagoats@gmail.com) Explanation of dataset: These data are model outputs used to populate Figure S1 in the paper. The raw data that went into the models were of ecological variables collected from a large-scale land-use and restoration experiment at Savannah River Site National Environmental Research Park in South Carolina, USA (33.20°N, 81.40°W). The experiment had a 2x2 factorial design resulting in 4 treatment combinations that are replicated across 27 blocks. The summary data are used to conduct a meta-analysis to understand how habitat restoration (tree removal) and land-use history (agriculture followed by plantation forestry) shape ecological attributes across 4 levels of organization: abiotic, organismal abundances, species diversity, and species interactions. Explanation of columns in the dataset: level - level of ecological property (abiotic, abundance, diversity, interactions) effect - four different effect sizes calculated from contrasts among the four treatment combinations. See figure 1 in the paper for illustration of the contrasts. "landuse.con" stands for "Ag history - Restoration, "landuse.thin" stands for "Ag. History + Restoration", "rest.pa" stands for "Restoration + Ag. History", and "rest.re" stands for "Restoration - Ag. History" type - categories for the type of variable measures (e.g. soils) or taxa (e.g. plants) estimate - Hedges g effect size estimate se - standard error of estimate zval - Z test statistic pval - P value ci.lb - lower bound of 95% confidence interval ci.ub - upper bound of 95% confidence interval