As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organised for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

ele1847-sup-0001-FigureS1-TableS1.docWord document1835KFigure S1 (pp. 2–3). Maps of individual county-level data sets used in the calculation of ROI: vertebrate species richness, estimated land cost ($/ha), percentage of county in protected areas and the percentage of natural land cover in a county threatened with conversion. For each variable mapped, the legend is divided into bins of 10% quantiles, with units as shown in the legend. Table S1 (p. 4). Total number of species ‘in need of protection’, area selected and estimated cost of land selected by three heuristics (full ROI, richness/cost and richness alone, see Methods) across 20 protection targets. ‘Target’ refers to the minimum percentage of every species’ range that must be in protected area (sum of existing protected areas and lands selected for acquisition by the heuristic), before a species is taken off the list in subsequent steps of the heuristic. Results for the three heuristics, from left to right under each heuristic title, are (1) the total area selected for acquisition across all counties (km2), (2) the cost of those lands based on the county-specific estimate of land costs, in $US billion ($B) and (3) the percentage of lands acquired that are considered overprotection, that is, lands that would not be threatened with conversion (see Fig. 1). In this article, Fig. 3 shows maps associated with protection targets 1, 7, 14 and 20% for the heuristic using the full ROI.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.